Introduction to Syntactic Annotation

Glossary
General principles Heads Modifiers Complements
Form vs. function: extended labels Case labels Function labels List of extended labels
PPCME2 Table of contents
YCOE Table of contents

This manual is meant to be used in conjunction with the PPCME2 manual. A basic understanding of the contents of the YCOE POS Manual and the PPCME2 Lite will also be helpful. When YCOE policy is the same as PPCME2 policy, a link to the PPCME2 manual is usually supplied without further comment. Additional information only applicable to the YCOE is supplied for each section, and significant differences from PPCME2 policy are flagged *difference*.

Glossary

The following terms have specific meanings within the YCOE annotation scheme. Words in boldface are defined.

ambiguous for case: a lexical form is ambiguous for case if it is formally ambiguous in isolation; dat/gen ambiguous means a form is formally identical in the dative and genitive
argument/complement: the terms argument and complement are used interchangeably
case-marked: case-marked refers to the addition of case in the annotation, not to any property of a lexical item; lexical items are referred to as (un)inflected, (un)declined, etc.
immediately dominate: a node immediately dominates another (or a word) if there is no intervening node between the two
extended label: an extended label is an addition to the first (usually formal) part of the label which is attached to it by a hyphen; there may be multiple extended labels (e.g., NP-NOM-PRD, CP-FRL-LOC, etc.)
label: a label is the set of letters following every open parenthesis, which indicates the type of constituent; a word-level label may also be referred to as a tag (NP-NOM ...)
level: a constituent is referred to as occurring or appearing at X-level, where X is the constituent that immediately dominates it (e.g. IP-level, phrasal level); level may also be used to refer to a class of constituents, as for example, word-level (also POS level) constituents
node: node is a way of referring to a set of labelled parentheses when thinking of it in terms of tree structure (a set of labelled parentheses represents a node in a tree structure); it is often used interchangeably with phrase
phrase/phrasal: used in a narrow sense to refer to the intermediate level of constituents between word/POS (N/ADJ/etc.) and clausal (IP/CP); NP, ADJP, ADVP, etc.; in a wider sense it refers to a set of labelled parentheses (especially in the term "project a phrase")
POS: part-of-speech; POS tags/labels are attached to the first set of parentheses surrounding a word (e.g., (PRO^N he))
specXP: where XP is some constituent category (specPP, specCP, etc.); this term is used rather loosely to refer to a position immediately preceding the head of the XP; it is a notational device within the YCOE/PPCME2 system and should not be interpreted in any particular case as having a linguistically valid analysis
tag: POS tag; sometimes also referred to as a label
token: a token is usually one matrix clause and all associated subordinate clauses, but it can be a smaller unit used independently; in the corpus each token is surrounded by a pair of empty parentheses and given an ID; a token is taken as an independent unit in all cases; relations are only indicated within tokens, never across tokens.

General Principles

Heads
Modifiers
Complements
PPCME2 Table of contents
YCOE Table of contents

The YCOE follows the same general principles for parsing as the PPCME2. The following additional principles apply only to the YCOE.

In general editor's punctuation is taken seriously as indicating preferred readings where there is a choice. This is especially the case when deciding whether to take a demonstrative pronoun as a relative or to start a new clause, but also applies in other situations, especially in the poetry. Subjects are given special status and are always distinguished from other arguments. Most commonly, subjects are indicated by the addition of the -NOM case tag, but when the subject is not (or would not be) nominative (as in the accusative-infinitive construction, non-nominative absolutes, and non-nominative small clauses), by a function tag -SBJ. In situ clausal subjects are also labelled -SBJ. Case and mood are generally indicated where they are unambiguous. Ambiguous forms are left unmarked. With the exception of nominative case which has special status (see Subjects), case percolates up from the part-of-speech tag to the syntactic labels. If the POS tag of a word is not marked for case (see Case), then the syntactic label will not carry case information either.

The internal syntax of all non-phrasal categories (i.e., excluding IP and CP) is fundamentally similar.

In summary:

The phrasal node (NP, PP, ADJP, etc.) immediately dominates the head category (N, P, ADJ, etc.); that is, there are no intermediate phrase-levels indicated.
Heads always project a phrasal node, with the following two exceptions:
1. certain heads, such as D (determiner), RP (particle) and all verbs, never project a phrase
2. prehead, single-word modifiers of nominal categories (i.e., N/Q/ADJ/NUM) do not project a phrasal node
3. posthead adjective and participle modifiers of nouns do not project a phrase in most cases (see Modifiers)
Complements always project a phrasal node, whether they consist of a single word or not.
Thus in the PPCME2/YCOE schema, both modifiers and complements are sisters of the head.

See the PPCME2 Lite for a full introduction to the annotation schema.

Heads

The head of a phrase is a word (POS) level category (N, Q, ADJ, etc.). In general the head of a phrase is of the same category as the phrase which dominates it (NP dominates N, etc.). As intermediate bar levels are not represented in the YCOE (see the PPCME2 Lite for an introduction to PPCME2/YCOE phrase structure), the phrase immediately dominates the head.

There are a number of cases, however, where a phrase dominates a "head" of a different category. In some cases these represent head categories (such as N) being divided into subcategories (e.g., PRO (pronoun) is a subcategory of N);


(NP-NOM (PRO^N he))

in others, the head may have been elided leaving a modifier behind, a process which we do not directly represent.


(NP-ACC (NP-ACC (ADJ^A fiftan) (N^A sunu))
	(CONJP (CONJ &)
	       (NP-ACC (ADJ^A syxtan))))           <--- SUNU elided

(NP-DAT-ADT (NP-DAT (ADJ^D Ebreisceon) (N^D stafon))
	    (CONJP (CONJ &)
		   (NP-DAT (ADJ^D Grecisceon)))    <--- STAFON elided
	    (CONJP (CONJ &)
		   (NP-DAT (NPR Leden) (N^D stafon))))

In most cases, however, we do not wish to make a decision about which of these analyses is the correct one. Thus, in general, when a phrase does not dominate a head of the same category, the user may view the phrase as one in which another word-level category acts as the head of a phrase, or alternatively as a phrase in which the head is omitted, elided, or non-overt for some reason.

Note that while a phrase may always be interpreted as having only one head (which may in some cases be non-overt), this does not mean that a phrase necessarily contains only one word-level category. See especially Modifiers.

The biggest disparity between word and phrase labels is found in NPs. NPs can be "headed" by:

pronouns (PRO, MAN) determiners (D) adjectives (ADJ) quantifiers (Q) numbers (NUM) participles (VBN, VAG, etc.)

As mentioned above, we are somewhat agnostic as to what the appropriate analysis of each of these cases is, but we divide them up as follows: determiners and pronouns are treated primarily as heads, while adjectives, quantifiers, numbers, and participles are treated primarily as modifiers. This means that when one of the latter category is modified, it projects its own phrase level, while the former, while seldom modified in any case, never do so; that is, determiners and pronouns are always immediately dominated by NP, while the other categories may have an intermediate phrase between them and the NP, and in any case, should always be interpreted as having such an intervening phrase (see modifiers for more detail on the rules governing the inclusion of a phrase label for modifiers). In general, a non-N category acts as head of an NP when it fills a nominal argument position (subject, object, etc.) or is the complement of another head.

As a consequence of this, in certain cases an NP may dominate two potential heads, a pronoun or determiner plus a quantifier or number. For the rules governing the modification of pronouns, see Posthead modification.


(NP-NOM (PRO^N hie) (Q^N ealle))

(NP-ACC (Q^A eal) (D^A +t+at))

(NP-DAT (PRO^D him) (NUM^D anum))

In general, NPs dominating only a QP are fairly common, while NPs dominating only an ADJP are very rare, and usually appear in elision contexts.


(NP-DAT (Q^D +anigum))                   <--- single Q dominated by NP

(NP-NOM (QP-NOM (ADV swa) (Q^N micel)))  <--- modified Q dominated by QP/NP

(NP-NOM (ADJ^N o+der))))                 <--- single ADJ dominated by NP

(NP-NOM (ADJP-NOM (ADV swy+de) (ADJ^N unforworhte))) <--- modified ADJ dominated by ADJP/NP

(PP (PP (P in)
	(NP-ACC (ADJP-ACC (ADV swa) (ADJ^A frecne))
		(N^A si+df+at)))
    (CONJP (CONJ &)
	   (PP (P in)
	       (NP-ACC (ADJP-ACC (ADV swa) (ADJ^A gewinfulne)))))   <--- elision
    (CONJP (CONJ &)
	   (PP (P in)
	       (NP-ACC (ADJP-ACC (ADV swa) (ADJ^A uncu+de))
		       (N^A el+teodignesse)))))

NPs immediately dominating two nouns (common or proper) are compounds.

ADJPs have a more limited range of possible "heads":

participles (VBN, VAG, etc.) participle phrases (PTP) possessive pronouns (PRO$) the quantifiers MICEL and LYTEL

In addition, numbers may occasionally appear to head ADJPs, but this is a special case always involving elision.

PPs are always headed by prepositions, although in some cases the preposition is cliticized to its complement (+T+ARON, etc.) and the POS tag includes both.


(PP (ADV+P +t+aron))

(PP (P+D^I +ar+tan))

ADVPs, apart from in the NEALLES +T+AT AN (+T+AT) construction, are always headed by adverbs.

QPs are always headed by quantifiers.

Number phrases (NUMP) are largely a notational convenience and do not always behave as other (linguistically valid) phrases do. NUMPs always contain one or more NUMs possibly along with conjunctions, all of which collectively act as the head, somewhat like in compounds.

Modifiers

Modifiers of nominal categories (NP/ADJP/QP/NUMP) are sisters to the head. Prehead single-word modifiers do not project a phrase-level since the relations between the two words in such a phrase are clear. In cases with a multi-word modifier (usually a modified modifier), a phrase-level is projected to make the relations clear. In the first example below, both adjectives modify the noun, and so both are sisters to it; in the second the adverb modifies the adjective (and thus is its sister) and the ADJP modifies the noun and is its sister.


(NP-DAT (ADJ^D o+drum) (ADJ^D langsumum) (N^D spr+acum)

(NP-ACC (ADJP-ACC (ADV swa) (ADJ^A deorwur+dne))
	(N^A cr+aft))

Modifiers of PPs are treated slightly differently; see Elements in specPP.

Posthead modifiers are treated somewhat differently.

any posthead modifier separated from the head by a comma projects a phrase
(NP-NOM (D^N +da) (ADJ^N wyrrestan) (, ,) (NP-NOM-PRN (ADJ^N faa) (N^N folcscea+dan)) (, ,) (NUMP-NOM (NUM^N feowertyne))) (NP-NOM (N^N licsyrce) (PRO$^N min) (, ,) (ADJP-NOM (ADJ^N heard)) (, ,) (ADJP-NOM (ADJ^N hondlocen)))

otherwise, with nominal heads, modifiers (adjectives/quantifiers/numbers/participles) are treated in the same way on both sides of the head (*difference* In the PPCME2 posthead modifiers always project a phrase)
(NP-ACC (NP-GEN (NPR^G Godes)) (N^A encgel) (ADJ^A haligne))

with pronominal heads,

adjectives and participles project a phrase in posthead position since they do not appear in prehead position as modifiers;
(NP-ACC-RFL (PRO^A +te) (ADJP-ACC (ADJ^A sylfne)))

quantifiers and numbers (which appear fairly freely on either side of pronouns) do not project a phrase
(NP-NOM (PRO^N hi) (Q^N ealle)) (NP-NOM (Q^N Ealle) (PRO^N hi)) (NP-ACC (PRO^A inc) (Q^A begen)) (NP-NOM (Q^N Sume) (PRO^N hi)) (NP-ACC (PRO^A +te) (NUM^A anne)) (NP-DAT (NUM^D anum) (PRO^D +te))

Modifiers are sometimes separated from a head with which they agree. It's not always clear in such cases, especially in the poetry, how such a modifier should be analyzed. Our general approach is to continue to label these elements as modifiers (i.e., as ADJP/QP/NUMP/PTP/etc., rather than as NP appositives, for instance) but not to trace them to the head as long as they are case-marked. This is meant to indicate an agnostic position as to the independence of these items. When not case-marked, they are traced to indicate the relationship. See also Separated modifiers.

(NODE (IP-MAT (NP-NOM (PRO^N Hi))
              (BEDI wurdon)
              (ADVP-TMP (ADV^T +ta))
              (QP-NOM (Q^N ealle))
              (PP (P +turh)
                  (NP-ACC (D^A +ta) (N^A wundra)))
              (RP+VBN^N onbryrde)
              (. ,))
      (ID AelfLives,+ALS_[Sebastian]:148.89)) 

(NODE (IP-MAT (CONJ &)
              (ADVP-LOC (ADV^L +t+ar))
              (BEDI w+as)
              (NP-NOM (NP-NOM (NPR^N Romane))
                      (CONJP *ICH*-1))
              (RP+VBN ofslagen)
              (NUMP-NOM (NUM eahtatig) (NUM^N +tusenda))
              (, ,)
              (CONJP-1 (CONJ &)
                       (NP-NOM (NP-NOM (PRO$ heora) (N^N consul))
                               (CONJP (CONJ &)
                                      (NP-NOM (PRO$ his) (NUM^N twegen) (N^N suna)))))
              (. .))
      (ID Orosius,Or_5:8.122.10.2479)) 

( (IP-MAT (NP-NOM (PRO^N We))
          (VBDI gefrunan)
          (PP (P on)
              (NP-DAT (N^D fyrndagum)))
          (NUMP-ACC (NUM^A twelfe))
          (PP (P under)
              (NP-DAT (N^D tunglum)))
          (NP-ACC (ADJ^A tireadige) (N^A h+ale+d)
                  (, ,)
                  (NP-ACC-PRN (NP-GEN (N^G +teodnes))
                              (N^A +tegnas)))
          (. .))
      (ID coandrea,3.1.5)) 

(NODE (IP-SUB-SPE (NP-NOM (PRO^N he))
                  (ADJP-1 (PRO$ his))      <--- non-case-marked ADJP traced
                  (ADJP-NOM (ADJ^N selfa))
                  (NEG ne)
                  (MDPI m+ag)
                  (PP (P $for)
                      (NP-DAT (PRO$ his) (N^D unsnyttrum)))
                  (NP-ACC (ADJP *ICH*-1)
                          (N^A ende))
                  (VB ge+tencean))
      (ID cobeowul,53.1728.1431)) 

( (CODE )
  (IP-MAT (NP-NOM (Q^N Sum) (N^N for+d+tegn)
                  (PTP *ICH*-1))
          (BEDI w+as)
          (ADVP-TMP (ADV^T +da))
          (ADJP-NOM-PRD (ADJ^N welig))
          (PP (P on)
              (NP-DAT (D^D +tam) (N^D lande)))
          (, ,)
          (PTP-1 (NP-NOM-PRD (NPR^N Florus))      <--- non-case-marked PTP traced
                 (VBN gehaten))
          (. ,)) (ID AelfLives,+ALS_[Maur]:125.356))

Complements

Unlike modifiers complements always have a phrase label, even when they are single words.


(NP (NP-GEN (NPR^G Cristes))
    (N geleafan)))

Note that while genitive NPs are treated like complements in all cases, possessive pronouns are usually treated as modifiers (but see also Possessive pronouns labelled NP for cases where possessives are labelled as genitive rather than possessive.)


(NP-NOM (PRO$ heora) (N^N magas))

At IP-level, NP complements of the verb are always distinguished from NP adjuncts (see Arguments vs. adjuncts); but note that this distinction is sometimes difficult to make and such distinctions should be verified before being used in any analysis for which this distinction is crucial. For PPs this distinction is not made as it is notoriously unreliable.

Within phrasal categories (except participle phrases which are essentially sentential in their internal syntax) the function of associated NPs is not indicated, apart from NPs of extent and appositives. NPs are simply labelled for case.

Within NPs NPs which are a sisters to the head are most commonly genitive (subjective/objective not distinguished), but occasionally dative. Dative/genitive ambiguous NPs in relation to an N are taken by default as genitive as long as this gives a posssible reading (see Case). Appositives on the head are distinguished by the -PRN label.


(NP-NOM (NP-GEN (D^G +tyssera) (N^G halgena))
	(N^N f+ader))

(NP-NOM (NPR^N Chromatius)
	(NP-NOM-PRN (D^N se) (N^N heahgerefa)))   <--- appositive

(NP-NOM (N^N wur+dmynt)
	(NP-DAT (D^D +tam) (ADJ^D welwyllendan) (NPR^D Gode)))

Similarly, within ADJPs and ADVPs the function of NPs is not distinguished apart from NP/QP modifiers indicating extent. NPs of extent also modify PPs, in which case they appear in specPP.


(ADVP-LOC (ADV^L gehende)
	  (NP-DAT (D^D +t+are) (NPR^D Liddan)))

(ADVP (NP-GEN (Q^G ealra))
      (ADV swi+dust))

(ADJP-NOM (ADJ^N uncu+d)
	  (NP-DAT (PRO^D him)))

(ADJP-NOM (NP-GEN (Q^G +alces) (N^G yfeles))
	  (ADJ^N orsorge)
	  (PP (P on)
	      (NP-DAT (ADJ^D ecere) (N^D blysse))))

Form vs. function: extended labels

Case labels
Function labels
List of extended labels
PPCME2 Table of contents
YCOE Table of contents

Extended labels are used to indicate additional aspects of form and function beyond the basic information about the category (NP, ADJP, etc) of the phrase. Formal aspects include primarily case, which is always the first extended label. Other extended labels indicate function.

Case labels

Case labels are included on every phrasal label that dominates a part-of-speech tag on which case is indicated.

(NP-ACC (D^A +done) (ADJ^A halgan) (N^A w+ar)

(NP-NOM (NUM^N twegen) (N^N gebro+dra)
        (ADJP-NOM (VBN^N +a+telborene)
                  (PP (P for)
                      (NP (N worulde)))))

(NP (NP-GEN (PRO$ his) (NPR^G drihtnes))
    (N +tenunge))

(NP (QP (Q ma)
        (PP (P +tonne)
            (NP (NUMP (NUM twa) (NUM hund)))))
    (NP-GEN (N^G punda)))

(NP (PRO$ his) (Q micclan) (N sige)

When case is not indicated on the POS tag, nominative case is still added to the phrase label if the phrase is the subject of a finite clause. Note that this applies only in finite clauses and not in non-finite clauses, such as infinitives and small clauses, where the subject/predicate is often not nominative. In non-finite clauses subjecthood is indicated by the -SBJ label, not case.


(NODE (IP-SUB (ADVP-LOC (ADV^L +ter))
              (NEG ne)
              (MDD mihte)
              (ADVP (NEG+ADV na))
              (NP-NOM (Q ma)
                      (NP-GEN (N^G muneca)))
              (VB wunian)
              (, ,)
              (PP (D^I +te)
                  (P l+as)
                  (CP-ADV (C +te)
                          (IP-SUB (NP-DAT (D^D +dam) (N^D gebro+drum))
                                  (NP-NOM (N^N bigleofan))
                                  (VBD ateorode)))))
      (ID AelfLives,+ALS_[Maur]:265.443)) 

(NODE (IP-SUB-SPE (ADVP-LOC (ADV^L +t+ar))
                  (BEPI beo+t)
                  (NP-NOM (NUM feowertig)))
      (ID Heptateuch,Gen:18.29.744)) 

( (IP-MAT (CONJ and)
          (ADVP (ADV eac))
          (NP-NOM (Q fela))
          (RP+VBDI wi+dcw+adon)
          (NP-DAT (NP-GEN (D^G +t+as) (N^G cyninges))
                  (N^D h+asum))
          (. ,))
      (ID AelfLives,+ALS_[Maccabees]:25.3027))

Function labels

The rest of the extended labels indicate the various functions of the constituents of the clause. Like case, functions that are marked on the POS tag (locative, directional, and temporal on adverbs) percolate up to the phrasal level. Function labels that are not word-based, however, are only used at the level at which the function is relevant. Thus temporal, locative and directional information is marked on adverbs at word-level and percolates up to the highest level, since these adverbs (THEN, THERE, HITHER, etc.) themselves contain this information. NPs being used temporally (THIS YEAR) on the other hand are only labelled as temporal at IP-level. Most function labels are only used at one level.


( (IP-MAT (CONJ and)
          (NP-NOM *con*)
          (VBDI het)
          (IP-INF (NP-ACC-SBJ (PRO^A hine))
                  (ADVP-TMP (ADV^T symble))
                  (BE beon)
                  (PP (P +atforan)
                      (NP (PRO$ his) (N gesih+de))))
          (. .)) (ID AelfLives,+ALS_[Sebastian]:13.11))

(NODE (IP-SUB-SPE (ADVP *T*-2)
                  (NP-NOM (PRO^N he))
                  (BEDI w+as)
                  (ADVP-TMP (ADVP-TMP (ADV^T gyrsand+ag))
                            (CONJP (CONJ &)
                                   (NP-ACC (D^A +tis) (ADJ^A +aran) (N^A d+ag)))))
      (ID Heptateuch,Gen:31.5.1238)) 

( (IP-MAT (CONJ &)
          (NP-NOM *con*)
          (NP-ACC (D^A +t+at))
          (VBDI (VBDI heold) (CONJ &) (VBD reahte))
          (NP-ACC-TMP (NP-ACC (NUM +treotteno) (N^A ger))
                      (CONJP (CONJ &)
                             (NP-ACC (NUM syx) (N^A mona+d)))
                      (CONJP (CONJ &)
                             (NP-ACC (NUM tyn) (N^A dagas))))
          (. .))
      (ID Bede1,Bede_1:13.54.24.340))

List of extended labels

IP/CP labels

Type of IP/CP (e.g.-MAT, -SUB, -CON, -INF, -THT, -ADV, etc.): All IPs and CPs in the corpus are labelled for type/function
Direct speech (-SPE): All IPs/CPs that are part of a sequence of direct speech have the label -SPE as the last extended label
Appositive or parenthetical (-PRN): Only marked at the top level

NP/QP/ADJP/NUMP labels

Case (-NOM, -ACC, -DAT, etc)
When case is labelled (according to our case-labelling conventions), it is labelled at all levels
Other labels
All other labels added to these phrases are added once at the relevant level and not to any embedded labels

-ADT (adjunct): added to non-argument NP/QPs at IP- level
-RFL (reflexive): added to NPs within the same clause coreferential with the subject (always first label following case)
-PRN (appositive): indicates appositive or parenthetical constituent; last label before -SPE
-TMP, -LOC, -DIR, -DIX (temporal, locative, directional): added at IP-level to NPs used temporally, locatively and directionally
-EXT (extent): added to NPs indicating extent with respect to adjectives, adverbs, quantifiers and PPs
-PRD (predicate): marks predicate phrases at IP-level (only applies to NP/QP/ADJP/NUMP predicates, not clausal or PP)
-SBJ (subject): added to non-nominative (including ambiguous) subjects of infinitives and absolutes (also added to clausal subjects in situ)
-LFD, -RSP (left-dislocated, resumptive): left-dislocated and resumptive pairs are labelled at IP-level (can also apply to other phrase types (CP, ADVP, etc.))
-VOC (vocative): vocative NPs
-EXL (exclamation): nominal exclamations (also used for clausal exclamations)

ADVP labels (-TMP, -LOC, -DIR, -DIX)

Only temporal, locative, and directional adverbs are labelled. The labels are lexically based and thus appear at all levels

PP labels

PPs are not labelled for function, except occasionally as appositives -PRN

IP/CP labels apply in the following order:


    IP/CP -type -PRN -SPE

NP/QP/ADJP/NUMP labels apply in the following order:


    NP -case -RFL -ADT/PRD
    NP -case -VOC/EXL
    QP -case -ADT/PRD
    NP -case -TMP/LOC/DIR
    ADJP/NUMP -case -PRD

Adjective Phrases (ADJP)

Complements of adjectives
NUM + WINTRE/GEARE (adjective)
Predicates
Possessive pronouns labelled ADJP
Non-predicate adjectives
PPCME2 Table of contents
YCOE Table of contents

PPCME2

For the rules governing ADJ(P)s as modifiers, see Modifiers.

Complements of adjectives

While argument NPs are distinguished from adjuncts at the IP-level by the -ADT label (see Arguments vs. adjuncts), this distinction is harder to make reliably in phrasal constituents, and we therefore do not attempt to do so. NPs and PPs within ADJPs will commonly be complements of the adjective, but this is not necessarily the case.


(NODE (IP-SUB-SPE (NP-NOM (PRO$ heora) (ADJ^N dyre) (N^N gold))
                  (NEG ne)
                  (BEPI bi+d)
                  (ADJP-NOM-PRD (NP-ACC (NEG+Q^A naht)) <--- NP complement of ADJ
                                (ADJ^N wur+d))
                  (PP (P wi+d)
                      (NP-ACC (D^A +ta) (RP+VBN^A fores+adan) (N^A ma+dmas))))
      (ID AelfLives,+ALS_[Swithun]:52.2429)) 

(NODE (IP-SUB (ADVP *T*-1)
              (NP-NOM (PRO^N heo))
              (VBDS l+age)
              (PP (P on)
                  (NP-DAT (N^D sl+ape)))
              (ADJP-NOM (ADJ^N hal)
                        (NP-DAT (Q^D eallum) (N^D limum)))) <--- NP adjunct of ADJ
      (ID AelfLives,+ALS_[Alban]:90.2377))

In copular clauses it can be particularly difficult to determine whether a dative NP is a complement of an adjectival predicate or adjunct (e.g., dative of interest). In the prose, genitives are always taken as complements. Examples of adjectives that take complements (genitive or dative) are FUL, GELIC, LICWUR+TE, GECWEME, GEHYRSUM, GEM+ATE, WYR+TE.


(NODE (IP-SUB (NP-NOM (PRO^N he))
              (BEDI w+as)
              (ADJP-NOM-PRD (NP-GEN (N^G bealwes))
                            (ADJ^N full)))
      (ID AelfLives,+ALS[Agnes]:394.770)) 

(NODE (IP-MAT-SPE (NP-NOM (PRO$^N Eower) (N^N word))
                  (BEPI syndon)
                  (ADJP-NOM-PRD (NP-DAT (N^D winde))
                                (ADJ^N gelice)))
      (ID AelfLives,+ALS[Agatha]:18.803))

In addition, superlative adjectives may take a genitive.


(IP-MAT (NP-NOM (NP-GEN (D^G +Tyses) (N^G cyninges))
		(N^N cwen))
	(BEDI w+as)
	(ADJP-NOM-PRD (ADJ^N forcu+tost)
		      (NP-GEN (N^G wifa)))
	(...))

In all other cases, datives are put at IP-level, but are not labelled as adjuncts.


(IP-MAT (CONJ and)
	(NP-NOM (PRO^N he))
	(ADVP (ADV for+ty))
	(ADJP-NOM-PRD (ADJ^N dyre))
	(BEDI w+as)
	(NP-DAT (PRO$ his) (N^D lareowe)
		(NP-DAT-PRN (NPR^D Benedicte))
		(CP-REL ...)))

( (IP-MAT (NP-NOM (PRO$ his) (N^N brydbedd))
          (NP-DAT (PRO^D me))
          (BEPI is)
          (ADJP-NOM-PRD (ADJ^N gearo))
          (ADVP-TMP (ADV^T nu))
          (ADVP-TMP (ADV^T iu))
          (PP (P mid)
              (NP-DAT (N^D dreamum)))
          (. .))
  (ID AelfLives,+ALS[Agnes]:42.522))

NUM + WINTRE/GEARE (adjective)

Adjectives ending in WINTRE/GEARE (ANWINTRE, TWELFWINTRE) meaning x years old, are somewhat problematic because the numbers are sometimes separated from the rest of the adjective. They are treated as follows. When all the numbers are attached to -WINTRE:


(ADJP (ADJ twelfwintre))

When all the numbers are separated:


(ADJP (NUM 4) (ADJ wintre))

(ADJP (NUMP (NUM 4) (CONJ and) (NUM 20))
      (ADJ wintre))

When some are attached and some are not, it is done as elision:


(ADJP (ADJP (NUM fif))
      (CONJP (CONJ &)
             (ADJP (ADJ sixtigwintre))))


(ADJP (ADJP (ADJ sixtigwintre))
      (CONJP (CONJ &)
             (ADJP (NUM fif))))

The same construction occurs with GEARE and NIHTE.

Possessive pronouns labelled ADJP

First and second person personal possessive pronouns, although tagged PRO$, are inflected as adjectives, and thus when they act as predicates they are labelled as ADJPs. In the poetry texts, this is also true of 3rd person possessives.


( (IP-MAT-SPE (NEG ne)
              (BEPI sind)
              (NP-NOM (PRO^N hi))
              (ADVP (NEG+ADV na))
              (ADJP-NOM-PRD (PRO$^N +dine))
              (. ;))
      (ID CathHomII,+ACHom_II,_11:99.243.2133))

In the prose texts, 3rd person possessives, when not single-word modifiers of a noun, are labelled as genitive pronouns.

Non-predicate adjectives

When adjectives are neither part of an NP nor acting as predicates, they are given a phrase label and appear at IP-level. Such adjectives may function in a number of ways (secondary predicate, result, etc.), or, especially in poetry, simply be separated modifiers but we do not make any distinctions among them. When case-marked they are not traced to the head (indicating their semi-independent status), but when no case-information is marked, they are traced to indicate the relationship.


( (IP-MAT (NP-DAT-RFL-ADT (PRO^D Him))
          (ADVP-TMP (ADV^T +da))
          (NP-NOM (NPR^N Scyld))
          (VBDI gewat)
          (PP (P to)
              (NP-DAT (N^D gesc+aphwile)))
          (ADJP-NOM (ADJ^N felahror))
          (VB feran)
          (PP (P on)
              (NP-ACC (NP-GEN (N^G frean))
                      (N^A w+are)))
          (. .)) (ID cobeowul,3.26.25))

( (IP-MAT (VBDI Aledon)
          (NP-NOM (D^N +ta))
          (NP-ACC (ADJ^A leofne) (N^A +teoden)
                  (, ,)
                  (NP-ACC-PRN (NP-GEN (N^G beaga))
                              (N^A bryttan)))
          (, ,)
          (PP (P on)
              (NP-ACC (N^A bearm) 
                      (NP-GEN (N^G scipes))))
          (, ,)
          (ADJP-ACC (ADJ^A m+arne))
          (PP (P be)
              (NP-DAT (N^D m+aste)))
          (. .)) (ID cobeowul,4.34.32))

( (IP-MAT (CONJ and)
          (NP-NOM (D^N se) (N^N mann))
          (ADVP-TMP (ADV^T sona))
          (ADJP-NOM (ADJ^N gesundful))
          (VBDI aras)
          (. ,)) (ID AelfLives,+ALS_[Sebastian]:359.218))

( (IP-MAT-SPE (NP-NOM (PRO^N Ic))
              (NP-DAT (PRO^D +te))
              (, ,)
              (NP-NOM-VOC (NPR^N Matheus))
              (, ,)
              (ADJP-ACC (PRO$^A mine))      <--- separated possessive pronoun
              (VBP sylle)
              (NP-ACC (N^A sybbe))
              (PP (P under)
                  (NP-DAT (N^D swegle)))
              (. .))
      (ID coandrea,5.94.68)) 

(NODE (IP-SUB-SPE (NP-NOM (PRO^N he))
                  (ADJP-1 (PRO$ his))      <--- non-case-marked ADJP traced
                  (ADJP-NOM (ADJ^N selfa))
                  (NEG ne)
                  (MDPI m+ag)
                  (PP (P $for)
                      (NP-DAT (PRO$ his) (N^D unsnyttrum)))
                  (NP-ACC (ADJP *ICH*-1)
                          (N^A ende))
                  (VB ge+tencean))
      (ID cobeowul,53.1728.1431))

Participle Phrase (PTP)

*difference*
There is no PTP label in the PPCME2. Phrases headed by participles are either adjunct participials (IP-PPL), absolutes (IP-ABS), or reduced relatives (RRC).

All phrases headed by a participle that is not part of the main verb sequence (i.e., not part of a periphrastic passive or perfect verb sequence) are labelled PTP. Reliably distinguishing the different functions of such phrases in all cases is extremely difficult, and thus PTP is a formal not a functional label and PTPs fulfill a number of different functions, including those covered in the PPCME2 by adjunct participials (IP-PPL), absolutes (IP-ABS), and reduced relatives (RRC)

PTPs inherit case from the participial head in the usual way. All participles in PTPs are marked for case except the "naming" participles (GE)HATEN, (GE)CIGEN, (GE)CWEDEN, and (GE)NAMEN (see Case on participles); likewise PTPs dominating such participles are not marked for case.


(NODE (IP-MAT (ADVP-TMP (ADV^T +Ta))
              (VBD +ateowde)
              (NP-NOM (NPR^N Sebastianus))
              (PP (P on)
                  (NP-ACC (N^A sw+afne)))
              (NP-DAT (NUM^D anre)
                      (N^D wudewan)
                      (, ,)
                      (PTP (NP-NOM-PRD (NPR^N Lucina))  <--- non-case-marked PTP
                           (VBN geciged))
                      (, ,)
                      (NP-NOM-PRN (ADJP-NOM (ADV swi+de) (ADJ^N +awf+ast))
                                  (N^N man)))
              (. ,))
      (ID AelfLives,+ALS_[Sebastian]:461.280))

PTPs are essentially sentential in their internal syntax. This means that unlike in ADJPs, for example, modifying adverbs have a phrasal label, adjunct NPs are labelled -ADT, etc. It also means that any NP without an ADT label within a PTP is an argument of the participle.


(ADJP-NOM (ADV swi+de) (ADJ^N halig))

(IP-MAT (NP-NOM (PRO^N he))
	(BEDI w+as)
	(ADVP (ADV yfele))
	(VBN^N geborene))

(PTP-NOM (ADVP (ADV yfele))
	 (VBN^N geborene))

(PTP-NOM (QP-DAT-ADT (Q^D micclum))    <--- adjunct
	 (NP-DAT (NPR^D Gode))         <--- argument
	 (VAG^N +tanciende))

PTPs which are contiguous to the NP which they modify are put inside that NP. If the the head of the NP is a noun, this applies to both preceding and following PTPs; when the head is a pronoun, however, only following PTPs are put inside, preceding PTPs are at IP-level. PTPs that are not contiguous to the NP they modify are treated like ADJPs and not traced as long as they are case-marked. If they are not case-marked, however, they are traced in order to make the relationship clear.


( (IP-MAT (CONJ ac)
          (BEDS w+are)
          (NP-NOM (D^N +t+at)
                  (N^N getel)
                  (PTP-NOM (VAG^N wunigende)  <--- PTP inside NP
                           (ADVP-TMP (ADV^T +afre))))
          (, ,)
          (NP-NOM-PRD (QP-NOM (NEG+CONJ ne) (Q l+as) (NEG+CONJ ne) (Q ma)))
          (, ,)
          (PP (P on)
              (NP-DAT (D^D +tam) (N^D munuclife)))
          (. .))
      (ID AelfLives,+ALS_[Maur]:265.444)) 

(NODE (IP-SUB (NP-NOM (D^N +da)
                      (N^N munecas)
                      (PTP-NOM (QP-DAT-ADT (Q^D micclum))
                               (VBN^N afyrhte)))
              (BEDI wurdon)
              (VBN^N awrehte)
              (PP (P +durh)
                  (NP (PRO$ his) (ADJ wodlican) (N stemne))))
      (ID AelfLives,+ALS_[Maur]:315.469)) 

(NODE (IP-MAT (NP-NOM (PRO^N Hi))
              (VBDI stodon)
              (ADVP-TMP (ADV^T +ta))
              (ADJP-NOM (ADJ^N stille))
              (PTP-NOM (PP (P on)              <--- PTP separated
                           (NP-DAT (D^D +tam) (N^D stocce)))
                       (VBN^N gef+astnode))
              (PTP-NOM (PP (PP (P ofer)
                               (NP-ACC (N^A d+ag)))
                           (CONJP (CONJ and)
                                  (PP (P ofer)
                                      (NP-ACC (N^A niht)))))
                       (NP-ACC (PRO$ heora) (NPR^A Drihten))
                       (VAG^N herigende))
              (. .))
      (ID AelfLives,+ALS_[Sebastian]:402.244)) 

(NODE (IP-MAT (CONJ &)
              (NP-NOM (PRO^N hi))
              (NP-ACC (PRO^A hine)
                      (PTP-ACC (VBN^A gebundenne)))  <--- PTP following pronoun
              (PP (P to)
                  (NP-DAT (PRO^D him)))
              (VBDI l+addon)
              (. .))
      (ID Bede1,Bede_1:7.34.28.110)) 

( (IP-MAT (CONJ &)
          (PTP-NOM (VAG^N re+dgiende))               <--- PTP preceding pronoun
          (NP-NOM (PRO^N hi))
          (NP-ACC (NPR^A Libertinum))
          (VBDI sohton)
          (. ,))
      (ID GDC1,:2.16.14.81)) 

(NODE (IP-MAT (PP (P On)
                  (NP-DAT (D^D +tam) (N^D dagum)))
              (BEDI w+as)
              (NP-NOM (Q^N sum)
                      (ADJ^N wis)
                      (N^N papa)
                      (NP-NOM-PRN *ICH*-1)
                      (PTP *ICH*-2))
              (PP (P on)
                  (NP (NPR Rome)))
              (, ,)
              (PTP-2 (NP-NOM-PRD (NPR^N Gaius))   <--- non-case-marked traced PTP
                     (VBN gehaten))
              (NP-NOM-PRN-1 (NP-GEN (ADJ^G haliges) (N^G lifes))
                            (N^N mann))
              (. .))
      (ID AelfLives,+ALS_[Sebastian]:321.195))

Adverb Phrases (ADVP)

Modifiers of adverbs
Adverbs modified by NP/QPs of extent
Adverbs taking NP complements
NEAH, GEHENDE, FEOR Other adverbs taking datives Superlative adverbs plus genitive
Subclasses of ADVPS (locative, temporal, directional)
ADVPs of extent
PPCME2 Table of contents
YCOE Table of contents

Modifiers of adverbs

PPCME2

It is sometimes difficult to decide when two adverbs appear in a row whether they form one or two constituents. This is especially true of temporal adverbs. The following list of adverbs are taken as modifying adverbs and are always grouped with a following adverb within a single ADVP.

EFNE, FOR, FUL, GENOH, HWENE, NA, SWA, SWI+TE, TO, +TUS, UNGEMET


(ADVP (ADV swa) (ADV w+alhreowlice))

(ADVP (ADV swa) (ADV +teah)) 

(ADVP (ADV for) (ADV wel))

(ADVP (ADV swy+de) (ADV rihtlice))

All other possible combinations are labelled separately when they are written separately. When written as one word they are labelled as one word. These include such common pairs as:

+TA +GIET, NU +GIET, +T+AR INN(E), +TA SONA


(NODE (IP-MAT (NP-NOM (PRO^N Hi))
              (ADVP-TMP (ADV^T +ta))
              (ADVP-TMP (ADV^T sona))
              (QP-NOM-NOM (Q^N begen))
              (VBDI begyrndon)
              (NP-ACC (PRO^A hi))
              (ADVP (ADV caflice))
              (. ,))
      (ID AelfLives,+ALS_[Sebastian]:247.147))

( (IP-MAT (CONJ &)
          (NP-NOM *con*)
          (MDD wolde)
          (VB faran)
          (ADVP-TMP (ADV^T +ta))
          (ADVP-TMP (ADV^T giet))
          (PP (P on)
              (NP-ACC (N^A herga+t)))
          (PP (P wi+d)
              (NP-GEN (NPR^G Ircingafeldes)))
          (. ;))
      (ID ChronA3,:918.7.744))

( (IP-MAT (ADVP-TMP (ADV^T nu))
          (ADVP-TMP (ADV^T iu))
          (NP-NOM (PRO^N ic))
          (BEPI eom)
          (VBN beclypt)
          (PP (P mid)
              (NP-DAT (PRO$ his) (ADJ^D cl+anum) (N^D earmum)))
          (. .))
      (ID AelfLives,AELS[Agnes]:45.525))

Adverbs modified by NP/QPs of extent

Adverbs can be modified by NP/QPs of extent.


(ADVP (NP-INS-EXT (D^I +ty))
      (ADV ea+d))

(ADVP (NP-INS-EXT (D^I +de))
      (ADV ra+dor))

Adverbs taking NP complements

NEAH, GEHENDE, FEOR

Although NEAH, GEHENDE, and FEOR act in some ways like prepositions in that they appear to take dative complements, they are also modified by such adverbs as SWA, SWI+DE, etc. in a non-prepositional way. We have therefore tagged these three words as adverbs when they do not appear as part of an NP or are not overtly inflected (in which case they are tagged as adjectives). A dative which goes with the adverb is included within the ADVP. The default labelling for cases where the expected inflection is zero is to take them as adverbs unless they occur within an NP. This includes the copular case. NEAH and GEHENDE are labelled as locative adverbs (apart from the use of NEAH to mean nearly), while FEOR may be locative or directional. FEOR is also used to indicate extent, in which case it is labelled -EXT.


( (CODE )
  (IP-MAT (ADVP-LOC (ADV^L +T+ar))
          (BEDI w+as)
          (ADVP (ADV eac))
          (VBN geset)
          (ADVP-LOC (ADV swi+te) (ADV^L gehende)
                    (NP-DAT (D^D +tam) (N^D mere)))
          (, ,)
          (NP-NOM (ADJ^N wearm) (N^N w+ater)
                  (PP (P on)
                      (NP (N cyfe))))
	  (...)))

( (IP-MAT (INTJ ge)
          (NP-NOM (Q^N ealle) (D^N +da) (N^N clifu)
                  (CP-REL (WNP-NOM-1 0)
                          (C +te)
                          (IP-SUB (NP-NOM *T*-1)
                                  (ADVP-LOC (ADV^L neah)
                                            (NP-DAT (D^D +t+am) (N^D s+a)))
                                  (BEDI w+aron))))
          (VBDI forburnan)
          (PP (P to)
              (NP (N ascan)))
          (. ;)) (ID Orosius,Or_5:4.119.9.2408))

(CP-THT (C +t+at)
        (IP-SUB (NP-NOM (D^N +ta) (N^N foreweardas))
		(BEDI w+aron)
		(ADVP-LOC (ADV^L feor)
			  (NP-DAT (D^D +d+am) (N^D f+astenne)))
		(VBN^N gesette)))

( (IP-MAT-SPE (CONJ &)
              (ADVP (ADV +teahhw+a+dere))
              (NEG ne)
              (VBP fare)
              (NP-NOM (PRO^N ge))
              (ADVP-DIR (ADV to) (ADV^D feor))
              (. :)) (ID Heptateuch,Exod:8.28.2666))

( (CODE )
  (IP-MAT (ADVP-TMP (ADV^T +Ta))
          (BEDI w+as)
          (NP-NOM (PRO^N he))
          (ADVP-LOC (ADVP-EXT (ADV swa) (ADV feor)
                              (PP *ICH*-2))
                    (ADV^L nor+t)
                    (PP-2 (P swa) 
                          (CP-CMPX (NP-NOM (D^N +ta) (N^N hw+alhuntan))
                                   (ADVP-DIR (ADV^D firrest))
                                   (VBPI fara+t))))
          (. .)) (ID Orosius,:1.14.9.130))

Other adverbs taking datives

Adverbs with an adjectival counterpart that takes dative (GELIC, GECWEMLIC, etc.) may also take dative complements.


(ADVP (NP-DAT (NPR^D Gode))
      (ADV gecwemlice))

(ADVP (NP-DAT (ADJ^D halgum))
      (ADV gelice))

Superlative adverbs plus genitive

Superlative adverbs commonly take genitive complements.


(ADVP (NP-GEN (Q^G ealra))
      (ADV swi+dust))

(ADVP-TMP (ADV^T +arest)
	  (NP-GEN (NPR^G Ongolcyninga)))

Subclasses of ADVPs (locative, temporal, directional)

A limited number of types of adverb are distinguished, locative (-LOC), temporal (-TMP), inherent directional (-DIX), and contextual directional (-DIR). These functions are marked on the POS tag, and percolate up to the phrase level.


( (IP-MAT (CONJ and)
          (NP-NOM *con*)
          (VBDI het)
          (IP-INF (NP-ACC-SBJ (PRO^A hine))
                  (ADVP-TMP (ADV^T symble))
                  (BE beon)
                  (PP (P +atforan)
                      (NP (PRO$ his) (N gesih+de))))
          (. .)) (ID AelfLives,+ALS_[Sebastian]:13.11))


( (CODE )
  (IP-MAT (ADVP-LOC (ADV^L +T+ar))
          (VBDI comon)
          (ADVP (ADV eac))
          (NP-NOM (PRO$ heora) (N^N magas))
          (. FULL-STOP)) (ID AelfLives,+ALS_[Sebastian]:40.25))

( (CODE )
  (IP-MAT (PP (P On)
              (NP-ACC (Q^A sumne) (N^A s+al)))
          (VBD eode)
          (NP-NOM (D^N se) (ADJ^N halga) (NPR^N Maurus))
          (ADVP-DIR (ADV^D ham))
          (PP (P to)
              (PP (NP-DAT (N^D mynstre))
                  (P weard)))
          (PP (P mid)
              (NP-DAT (PRO$ his) (N^D gebro+drum)))
          (. .)) (ID AelfLives,+ALS_[Maur]:15.296))

( (IP-MAT (CONJ and)
          (NP-NOM (PRO^N hi))
          (ADVP-DIX (ADV^DX +tyder))
          (VBDI comon)
          (PP (P mid)
              (NP-DAT (Q^D mycelre) (N^D sarnyssa)))
          (CP-FRL-DIR (WADVP-DIR-1 (ADV^D +t+ar))
                      (C 0)
                      (IP-SUB (ADVP-DIR *T*-1)
                              (NP-NOM (PRO$ heora) (N^N suna))
                              (BEDI w+aron)
                              (VBN^N geh+afte)))
          (. ,)) (ID AelfLives,+ALS_[Sebastian]:34.23))

ADVPs of extent

The adverb FEOR may be used to indicate extent like NPs of extents indicating distance.


(ADVP-DIX (ADVP-EXT (NEG+Q naht) (ADV feor))
	  (ADV^DX +tanon))

(PP (ADVP-EXT (NEG+Q naht) (ADV feor))
    (P fram)
    (NP-DAT (PRO$ his) (N^D mynstre)))

(ADVP-DIR (ADVP-EXT (ADV feorr))
	  (ADV^D ofdun))

(PP (ADVP-EXT (ADV swa) (ADV feor)) 
    (P fram)
    (NP-DAT (PRO^D him)))

Quantifier Phrase (QP)

Quantifiers as heads or modifiers
Floated quantifiers
Adverbial use of quantifiers
Uninflected EALL
PPCME2 Table of contents
YCOE Table of contents

PPCME2

*difference*
In the PPCME2 QPs do not occur at IP-level; rather they are always contained within an NP. In YCOE floated quantifiers and quantifiers used adverbially occur at IP-level. Uninflected EALL is a special case.

Modifying quantifiers appear on both sides of the head, even with pronouns (unlike adjectives), and as such single quantifiers do not project a phrase in either position with either nomininal or pronominal heads. See Modifiers for examples.

Quantifiers as heads or modifiers

Like numbers, quantifiers either modify nominals (i.e. agree in case) or take genitive complements. For ambiguous cases, the rule is that the undeclined quantifiers FELA, MA, and LYT take genitives, while all others are taken as modifiers. This follows the majority pattern, although there are clear cases which go against it in both directions.


(NP (Q ma)
    (NP-GEN (N^G suna)))

(NP-ACC (Q fela)
	(NP-GEN (VAG|G byrnenda) (N|G gleda))) 

(NP-NOM (Q^N manige) (N^N suna))

(NP-NOM (Q^N ealle) (PRO$ hire) (N^N wunda))

(NP-NOM (Q Fela) (ADJ^N dyslice) (N^N d+ada)) <--- FELA agreeing

Floated quantifiers

Floated quantifiers are labelled QP-case but are not traced to their head noun.


( (CODE )
  (IP-MAT (NP-NOM (PRO^N Hi))
          (BEDI wurdon)
          (ADVP-TMP (ADV^T +ta))
          (QP-NOM (Q^N ealle))
          (PP (P +turh)
              (NP-ACC (D^A +ta) (N^A wundra)))
          (RP+VBN^N onbryrde)
          (. ,)))

Adverbial use of quantifiers

Inflected quantifiers used adverbially are labelled QP-case-ADT. Uninflected quantifiers like MA, LYT, FELA and NAHT are labelled QP-ADT. The quantifier may be bare or modified.


(QP-GEN-ADT (Q^G ealles))

(QP-DAT-ADT (ADV swa) (Q^D mycclum))

(QP-ADT (Q ma))

(QP-ADT (NEG+Q naht))


( (CODE )
  (IP-MAT (ADVP-TMP (ADV^T +Ta))
          (RP+AXDI ongunnon)
          (NP-NOM (PRO$ heora) (N^N magas))
          (QP-DAT-ADT (Q^D mycclum))
          (VB behreowsian)
          (CP-THT (C +t+at)
                  (IP-SUB (NP-NOM (PRO^N hi))
                          (ADVP-TMP (ADV^T +afre))
                          (NP-ACC (D^A +ta) (N^A martyras))
                          (VB misl+aran)
                          (MDDI woldon)))
          (. ,)))

( (CODE )
  (IP-MAT (ADVP-TMP (ADV^T +Ta))
          (NEG ne)
          (MDD dorste)
          (NP-NOM (D^N se) (N^N heahgerefa))
          (QP-ADT (NEG+Q naht))
          (PP (P ongean)
              (NP-ACC (D^A +ta) (N^A h+a+dengyldan)))
          (. ,)))

Uninflected EALL

When EALL is uninflected it may either modify a noun in the masc.nsg. or neut. n/asg. or be used adverbially. It is extremely difficult to distinguish these two cases in some instances. We have adopted the following strategy for these cases.

EALL immediately preceding a head with which it could agree, is taken as modifying that head (IP-SUB-0 (NP-NOM (PRO^N he) (ADJP-NOM (ADJ^N sylf))) (MDD wolde) (, ,) (PP (P gif) (CP-ADV (C 0) (IP-SUB (NP-ACC (PRO^A hit)) (NP-NOM (NPR^N God)) (ADVP (ADV swa)) (RP+VBD foresceawode)))) (NP-ACC (Q^A eall) (N^A woruld+ding)) (VB forl+atan)) EALL in all other positions (including immediately following a head with which it could agree) is labelled QP and put at IP-level. Note that the QP is not further specified with an -ADT label in order to indicate its unclear status. ( (CODE ) (IP-MAT (PP (P +After) (NP-DAT (D^D +tysum) (N^D worde))) (NP-NOM (PRO^N he)) (BEDI wear+d) (QP (Q eall)) (VBN geh+aled) (. ,)) (ID AelfLives,+ALS_[Sebastian]:299.181)) ( (IP-MAT (CONJ and) (NP-NOM *con*) (VBD ar+arde) (NP-ACC (D^A +t+at) (N^A mynster)) (QP (Q eall)) (PP (P be) (NP-DAT (NP-GEN (NPR^G Maures)) (N^D r+ade))) (PP (P wi+d) (NP-ACC (D^A +ta) (Q^A mycclan) (N^A ea) (CP-REL (WNP-1 0) (C +te) (IP-SUB (NP-NOM (N^N menn)) (VBPI hata+d) (IP-SMC (NP-SBJ *T*-1) (NP-PRD (NPR Liger))))))) (. .)) (ID AelfLives,+ALS_[Maur]:157.368))

Number Phrase (NUMP)

Number as head or modifier
The number AN
TWEGRA ELNA BRAD etc.
Complex number expressions (ONE LESS (THAN) FORTY, MORE THAN 300, etc.)
NUMPs as predicates
Separated numbers
PPCME2 Table of contents
YCOE Table of contents

PPCME2

A number phrase consists of either a multi-word number or a number modified in some way. The NUMP label is used largely as a convenience to group together single numbers when they are written as multiple words. In this case NUMP has little if any linguistic validity and does not follow the usual rules for phrases (General principles). Rather all the members of such a NUMP are sisters. Numbers are sometimes modified, however, (almost ten, about thirty, etc.) and such modifiers are included in the NUMP. As usual, single-word modifiers do not have a phrasal label but multi-word modifiers do.


(NUMP (NUM feower) (CONJ and) (NUM twentig))

(NUMP (ADV wel) (NUM hundseofontig))

(NUMP-ACC (ADV huru) (NUM^A +dry))

Like other modifiers, single-word numbers separated from their head by a comma are dominated by NUMP.


( (IP-MAT (CONJ ach)
          (NP-GEN-1 (D^G +t+as) (N^G weorodes))
          (ADVP (ADV eac))
          (NP-NOM (NP-GEN *ICH*-1)
                  (D^N +da)
                  (ADJ^N wyrrestan)
                  (, ,)
                  (NP-NOM-PRN (ADJ^N faa) (N^N folcscea+dan))
                  (, ,)
                  (NUMP-NOM (NUM^N feowertyne)))   <--- only ex. in corpus
          (AXDI gewiton)
          (PP (P mid)
              (NP-INS (D^I +ty) (N^I w+age)))
          (PP (P in)
              (NP-ACC (N^A forwyrd)))
          (VB sceacan)
          (PP (P under)
              (NP-ACC (NP-GEN (N^G eor+tan))
                      (N^A grund)))
          (. .))
      (ID coandrea,47.1591.661))

When a numeral is separated by more than punctuation from the noun it modifies, it is put at clause level with a NUMP label (see separated modifiers).


( (IP-MAT (NP-NOM (PRO^N We))
          (VBDI gefrunan)
          (PP (P on)
              (NP-DAT (N^D fyrndagum)))
          (NUMP-ACC (NUM^A twelfe))       <--- number
          (PP (P under)
              (NP-DAT (N^D tunglum)))
          (NP-ACC (ADJ^A tireadige)       <--- agrees with
                  (N^A h+ale+d)
                  (, ,)
                  (NP-ACC-PRN (NP-GEN (N^G +teodnes))
                              (N^A +tegnas)))
          (. .))
      (ID coandrea,3.1.5))

When split, multi-word numbers are treated as cases of single-word conjunction. This follows from the fact that NUMP is just a notational convenience to group numbers.


(NP-NOM (NUMP (NUM twentig) (NUM +tusend)
	      (CONJP *ICH*-2))
	(NP-GEN (N^G manna))
	(CONJP-2 (CONJ and) (NUM six) (NUM $hundred)))

Number as head or modifier

Numbers either modify nouns or take nominal complements in the genitive. Note in particular the position of the genitive complement (NP-GEN) in relation to the NUMP when present. The NUMP in this case is simply a notational device to group the number and the whole number is really the head, just as in the example with a single-word number.


(NP-ACC (NUM^A anne) (N^A stan))

(NP-NOM-PRD (NUMP (NUM twa) (CONJ and) (NUM hundseofontig)) 
	    (N^N geare))

(NP-NOM (NUM^N +try) 
	(NP-GEN (D^G +t+ara) (N^G wyrhtena)))

(NP-NOM (NUMP (NUM an) (NUM hund))     <--- NUMP and NP-GEN sisters
	(NP-GEN (N^G muneca)))

When a numeral without overt case is followed by a noun in -A which could be gen.pl. or nom/acc.pl (i.e. in particular feminine nouns and masculine nouns of the u-declension), the following policy is used, which uses the majority pattern as a default.

if the number is 1 to 19, the structure is parsed as a nom/acc.pl. NP:
```
(NP-NOM (NUM siex) (N^N suna))
```
if the number is 20 or higher, the noun in -A is assumed to be genitive:
```
(NP (NUM +tritig)
    (NP-GEN (N^G suna)))
```

The number AN

Note that AN is always treated as a number, even when it means alone. But see Focus particles for weak ANA used as a focus particle.


( (IP-MAT-SPE (PP (P gif)
                  (CP-ADV-SPE (C 0)
                              (IP-SUB-SPE (NP-NOM (PRO^N +tu))
                                          (NP-ACC (PRO^A hi))
                                          (RP+VBPI onscunast))))
              (NP-NOM (PRO^N wit))
              (VBPI cwe+da+t)
              (ADVP-TMP (ADV^T +tonne))
              (NUMP-NOM (NUM^N an))
              (. .))
      (ID AelfLives,+ALS[Agatha]:77.844))

TWEGRA ELNA BRAD etc.

The structure of phrases like TWEGRA ELNA BRAD two els wide is for the numbers 1 to 19:


(ADJP (NP-GEN-EXT (NUM^G twegra) (N^G elna))
      (ADJ brad))

(ADJP (NP-GEN-EXT (NUM nigon) (N^G elna))
      (ADJ brad))

and for 20 or higher:


(ADJP (NP-EXT (NUM twentig)
	      (NP-GEN (N^G elna)))
      (ADJ brad))

For an explanation of the -EXT label, see NPs of extent.

Complex number expressions (ONE LESS (THAN) FORTY, MORE THAN 300, etc.)

See pending issues.

NUMPs as predicates

In general NUMPs are dominated by NPs at clause level, but as predicates they are labelled NUMP. Note that this only applies to bare numbers and not to numbers with genitive complements, which are labelled as NPs. If the number is case-marked, it is labelled NUMP-NOM-PRD. Non-case-marked numbers are simply labelled NUMP-PRD (see Predicates).


( (CODE )
  (IP-MAT (ADVP (ADV So+dlice))
          (NP-NOM (D^N +ta)
                  (CP-REL (WNP-NOM-1 0)
                          (C +de)
                          (IP-SUB (NP-NOM *T*-1)
                                  (ADVP-LOC (ADV^L +tar))
                                  (VBDI +aton))))
          (BEDI w+aron)
          (NUMP-NOM-PRD (NUM fif) (NUM +tusend))
          (. ;)) (ID WSGospels,Mk_[WSCp]:8.9.2741))


( (IP-MAT (NP-NOM *pro*)
          (NUMP-NOM-PRD (NUM^N Twelfe))
          (BEDI $w+aron)
          (, ,)
          (ADJP-NOM (NP-DAT (N^D d+adum))
                    (ADJ^N domf+aste))
          (, ,)
          (PTP-NOM (NP-DAT-ADT (NPR^D dryhtne)) 
                   (VBN^N gecorene))
          (, ,)
          (ADJP-NOM (ADJ^N leofe)
                    (PP (P on)
                        (NP-DAT (N^D life))))
          (. .)) (ID cocynew,51.4.8))

Separated numbers

When the number phrase modifies an NP from which it is separated, like other modifiers it is labelled NUMP-case, if appropriate; if it is case-marked it is not traced to its head.


(NODE (IP-MAT (CONJ &)
              (ADVP-LOC (ADV^L +t+ar))
              (BEDI w+as)
              (NP-NOM (NP-NOM (NPR^N Romane))
                      (CONJP *ICH*-1))
              (RP+VBN ofslagen)
              (NUMP-NOM (NUM eahtatig) (NUM^N +tusenda))
              (, ,)
              (CONJP-1 (CONJ &)
                       (NP-NOM (NP-NOM (PRO$ heora) (N^N consul))
                               (CONJP (CONJ &)
                                      (NP-NOM (PRO$ his) (NUM^N twegen) (N^N
 suna)))))
              (. .))
      (ID Orosius,Or_5:8.122.10.2479)) 

( (IP-MAT (NP-NOM (PRO^N We))
          (VBDI gefrunan)
          (PP (P on)
              (NP-DAT (N^D fyrndagum)))
          (NUMP-ACC (NUM^A twelfe))
          (PP (P under)
              (NP-DAT (N^D tunglum)))
          (NP-ACC (ADJ^A tireadige)
                  (N^A h+ale+d)
                  (, ,)
                  (NP-ACC-PRN (NP-GEN (N^G +teodnes))
                              (N^A +tegnas)))
          (. .))
      (ID coandrea,3.1.5)) 

( (IP-MAT (NUMP-NOM (NUM^N Fife))
          (VBDI l+agun)
          (PP (P on) 
	      (NP-DAT (D^D +tam) (N^D campstede)))
          (NP-NOM (NP-NOM (N^N cyningas)
                          (ADJ^N giunge)
                          (, ,)
                          (PTP-NOM (NP-DAT (N^D sweordum))
                                   (VBN^N aswefede)))
                  (, ,)
                  (CONJP (ADV swilce)
                         (NUMP-NOM (NUM^N seofene))
                         (ADV eac)
                         (NP-NOM (N^N eorlas)
                                 (NP-GEN (NPR^G Anlafes))))
                  (, ,)
                  (CONJP (NP-NOM (N^N unrim)
                                 (NP-GEN (N^G heriges)
                                         (, ,)
                                         (NP-GEN-PRN (N^G flotan) (CONJ and) (NPR^G Sceotta))))))
          (. .))
  (ID cobrunan,18.28.21))

Noun Phrases (NP)

Prehead modification of pronouns
Posthead modification
Separated modifiers
Possessive pronouns labelled NP
Appositives

Compound nouns
Compound Proper Names Place names Personal names Temporal names
Clausal syntax of Noun Phrases
Arguments vs. adjuncts Subjects Subjects in tensed clauses Subjects in non-tensed clauses Subjects in poetry Presentational THERE Expletive IT Other arguments Reflexives (-RFL) Foreign words and phrases Dative of possession Adjunct NPs Temporal, locative and directional NPs Left-dislocated NPs
PPCME2 Table of contents
YCOE Table of contents

PPCME2

Prehead modification of pronouns

In general pronouns are not premodified except by quantifiers, but there are a small number of cases of premodification with SELF and numbers. Like all prehead modification modifiers in this position do not project a phrase.


( (IP-MAT (CONJ &)
          (NP-NOM (ADJ^N selfe) (PRO^N hy))
          (VBDI forwurdan)
          (. .))
      (ID WHom,WHom_20.3:184.1621)) 

(PP (P of)
    (NP-DAT (ADJ^D sylfum) (PRO^D +te)))

( (IP-MAT (NP-NOM (ADJ^N Deor)
                  (ADJP-NOM (NP-GEN (N^G domes))
                            (ADJ^N georn))
                  (, ,)
                  (PRO^N hio)
                  (ADJP-NOM (ADJ^N dumb)))
          (VBPI wuna+d)
          (. ;))
  (ID coriddle,196.16.452))

Posthead modification

With two exceptions, modifiers have equal status whichever side of the head they appear on. Thus single adjectives appearing before and after the noun are treated the same and not given a phrase label. This applies to quantifiers and numbers as well as adjectives. See also Modifiers.


(NP-NOM (ADJ^N adj1) (N^N noun) (ADJ^N adj2))

*difference*
Note that this differs from the PPCME2 where posthead modifiers are always given a phrasal tag.

The first exception to this policy is when the head is a pronoun. In this case following single adjectives are given a phrase label.


(NP-NOM (PRO^N pronoun)
	(ADJP-NOM (ADJ^N adj)))


(NODE (IP-MAT-SPE (NP-NOM-x *exp*)
                  (NP-NOM-PRD (N^N Wa))
                  (BEPI is)
                  (NP-DAT (PRO^D me)
                          (ADJP-DAT (ADJ^D earmum)))  <--- post-pronoun adjective
                  (CP-THT-SPE-x (C +t+at)
                                (IP-SUB-SPE (NP-NOM (PRO^N ic))
                                            (BEPI eom)
                                            (RP+VBN oferswy+ted)
                                            (PP (P fram)
                                                (NP-DAT (D^D +tysum) (ADJ^D halgum) (N^D werum))))))
      (ID AelfLives,+ALS[Forty_Soldiers]:222.1393))

The second exception is a following adjective (or other modifier) separated from the head by a comma in the text. This occurs largely in the poetry and is meant to indicate our uncertainty as to the status of these constituents


(NP-NOM (N^N licsyrce) (PRO$^N min)
	(, ,)
	(ADJP-NOM (ADJ^N heard))
	(, ,)
	(ADJP-NOM (ADJ^N hondlocen)))

Separated modifiers

When a modifier is separated from a head with which it agrees, it is given a phrase label and appears at IP-level. It is case-marked to agree with the head unless it is one of the categories that is never case-marked, in which case it is traced. This is most common in poetry, but in prose occurs with floated quantifiers and adjectives in various constructions (secondary predicate, result, etc.) which we do not distinguish. Predicate adjectives are marked as predicates.


( (CODE )
  (IP-MAT (NP-NOM (PRO^N Hi))
          (BEDI wurdon)
          (ADVP-TMP (ADV^T +ta))
          (QP-NOM (Q^N ealle))
          (PP (P +turh)
              (NP-ACC (D^A +ta) (N^A wundra)))
          (RP+VBN^N onbryrde)
          (. ,)) (ID AelfLives,+ALS_[Sebastian]:148.89))

( (IP-MAT (CONJ and)
          (NP-NOM (D^N se) (N^N mann))
          (ADVP-TMP (ADV^T sona))
          (ADJP-NOM (ADJ^N gesundful))
          (VBDI aras)
          (. ,)) (ID AelfLives,+ALS_[Sebastian]:359.218))

Separated possessive pronouns are labelled as ADJPs. Those that are case marked (non-third person) are treated as other modifiers and not traced.

In the poetry texts, third-person possessives are also labelled as ADJPs, but without case, and are traced to their head to make relations clear. See also Possessive pronouns labelled ADJP. In prose 3rd person possessive pronouns are treated as nominal in all cases except as single-word modifiers.


( (IP-MAT-SPE (NP-NOM (N^N Bl+ad))
              (BEPI is)
              (VBN ar+ared)
              (PP (P geond)
                  (NP-ACC (N^A widwegas)))
              (, ,)
              (NP-NOM-VOC (N^N wine) (PRO$^N min)
                          (NP-NOM-PRN (NPR^N $Beowulf)))
              (, ,)
              (ADJP-NOM (PRO$^N +din))  <--- separated possessive
              (PP (P ofer)
                  (NP-ACC (NP-GEN (N^G +teoda))
                          (Q^A gehwylce)))
              (. .))
      (ID cobeowul,52.1703.1405)) 

( (IP-MAT-SPE (NP-NOM (PRO^N Ic))
              (NP-DAT (PRO^D +te))
              (MDPI sceal)
              (ADJP-ACC (PRO$^A mine))  <--- separated possessive
              (VB gel+astan)
              (NP-ACC (N^A freode))
              (, ,)
              (PP (P swa)
                  (CP-CMPX-SPE (NP-NOM (PRO^N wit))
                               (ADVP-TMP (ADV^T fur+dum))
                               (VBDI spr+acon)))
              (. .))
      (ID cobeowul,52.1706.1409)) 

(NODE (IP-SUB-CAR-PRN (NP-NOM (PRO^N ic))
                      (ADJP-1 (PRO$ his))  <--- separated 3rd pers. possessive
                      (ADVP-TMP (ADV^T +arest))
                      (NP-DAT (PRO^D +de))
                      (NP-ACC (ADJP *ICH*-1)  <--- trace
                              (N^A est))
                      (VBD ges+agde)
      (ID cobeowul,67.2152.1761))

Possessive pronouns labelled NP

When single words modifying nouns, 3rd person possessives like other possessive pronouns are labelled PRO$, but without case since they do not inflect. In other positions they are labelled as genitive pronouns (PRO^G). This applies in four cases: when they are used as complement genitives, usually to quantifiers, when used as predicates, when conjoined with a genitive NP, and when modified by SELF.


(NP-NOM (NP-GEN (PRO^G heora))    <--- genitive complement of quantifier
	(Q^N +ag+der))

(IP-SUB (NP-GEN-2 (PRO^G his))
	(NEG ne)
	(MDD mihte)
	(NP-NOM (NEG+Q^N nan) (N^N man))
	(VB witan)
	(NP-ACC (NP-GEN *ICH*-2)
		(NEG+Q^A nan) (N^A gemet)))))

( (IP-MAT-SPE (CONJ &)
              (CP-FRL-SBJ-SPE (WNP-NOM-1 (D^N +d+at))
                          (C 0)
                          (IP-SUB-SPE (NP-NOM *T*-1)
                                  (ADVP-LOC (ADV^L +t+ar))
                                  (ADJP-NOM-PRD (ADJ^N dead))
                                  (BEPI bi+d)))
              (BEPI bi+t)
              (NP-GEN-PRD (PRO^G his))   <--- predicate HIS
              (. .)) (ID Heptateuch,Exod:21.33.3270))

(NP (NP-GEN (NP-GEN (PRO^G his))         <--- conjoined with genitive NP
	    (CONJP *ICH*-1)) 
    (N heortan)
    (CONJP-1 (CONJ &)
	     (NP-GEN (PRO$ his) (N^G folces))))

(NP-NOM (NP-GEN (PRO^G his)              <--- modified by SELF
		(ADJP-GEN (ADJ^G sylfes)))
	(N^N willa))

Appositives

PPCME2

In cases like Mary told about John how he... where there are two complements seemingly filling the same single complement position of the verb, the second complement is treated separately from the first, but also traced to it as an appositive to indicate that in some sense it is not a separate argument. This contrasts with normal clausal appositives which are coreferential with the noun they are appositive on.


(NODE (IP-MAT (CONJ Ac)
              (NP-NOM (PRO^N we))
              (ADVP-TMP (ADV^T nu))
              (MDPI sculon)
              (ADVP (ADV hr+adlice))
              (NP-ACC (Q^A hw+athwugu)
                      (CP-QUE-PRN *ICH*-1))
              (PP (P be)
                  (NP-DAT (PRO$ hire) (N^D for+dfore) (NUM^D anre)))
              (VB secgan)
              (, ,)
              (CP-QUE-1 (WADVP-2 (WADV hu))
                        (C 0)
                        (IP-SUB (ADVP *T*-2)
                                (NP-NOM (PRO^N heo))
                                (NP-ACC (D^A +t+at) (ADJ^A heofonlice) (N^A rice))
                                (VBD gesohte)))
              (. .))
      (ID Bede3,Bede_3:6.172.28.244)) 

(NODE (IP-SUB (NP-NOM (PRO^N he))
              (NP-ACC (D^A +ta) (ADJ^A eadigan) (N^A herenesse)
                      (CP-QUE-PRN *ICH*-1))
              (ADVP (ADV eac))
              (VBD gehyrde)
              (, ,)
              (CP-QUE-1 (WADVP-2 (WADV hu))
                        (C 0)
                        (IP-SUB (ADVP *T*-2)
                                (NP-NOM (PRO^N heo))
                                (NP-ACC (NPR^A God))
                                (VBDI (VBDI lofodon) (CONJ &) (VBDI heredon)))))
      (ID Bede3,Bede_3:14.212.6.687)) 

( (IP-MAT (NP-NOM *con*)
          (VBD anddette)
          (NP (PRO$ his) (N synne)
	      (CP-THT-PRN (C +t+at)      <--- normal clausal appositive
			  (IP-SUB (NP-NOM (PRO^N he))
				  (RP+VBDI ofsloh)
				  (NP-ACC (D^A +da) (ADJ^N halgan))))))
  (. ,)) (ID AelfLives,AELS[Agnes]:424.789))

Compound Nouns

Although most compounds are written as single words in the Old English texts, occasionally they are written as two separate words with the first part uninflected. These are bracketed flat with case indicated only on the second word. Any NP containing more than one N (with no conjunction) is a compound.


(NP-DAT (D^D +d+are) (N woruld) (N^D lufe))

(NP-ACC (N winter) (N^A setl))

(NP-ACC (N gang) (N^A dagas))

(NP-DAT (D^D +t+are) (N nor+t) (N^D s+a))

Compound Proper Names

The following sections apply only to the annotation of prose texts. In the poetry, all names are case-marked and parsed accordingly.

Place names

Place names of the type EGYPTA LOND, SEIR DUNE, ELIG MYNSTER, with a proper first part and a common second part are treated as compounds and done as follows. Case is never marked on the first part even when it is clear. The second part is labelled N-case following the usual case rules.


(NP-DAT (NPR Elig) (N^D mynstre))

(NP-DAT (NPR Antiochian) (N^D byrig))

(NP-ACC (NPR Egypta) (N^A lond))

(NP-DAT (NPR Iudea) (N^D lande))

Proper names themselves can also be compounds, usually including EAST, WEST, etc. These are true compounds (unlike at least some of the previous category) since the first part is never inflected.


(NP-GEN (NPR West) (NPR^G Seaxna))

Cases with a determiner like ROME +T+ARE BYRIG are always done as appositives, including when both parts are genitive and one could be taken as dependent on the other (i.e., they are always interpreted as ROME, THE CITY rather than THE CITY OF ROME).


(NP-DAT (NPR^D Rome)
	(NP-DAT-PRN (D^D +t+are) (N^D byrig)))

Personal names

Personal names preceded, or more often followed, by an office or title such as CYNING, BISCEOP, MUNUC, ABBOD, etc. are parsed flat, but both parts are case-marked as appropriate (unlike with place names).


(NP-DAT (NPR^D +Alfstane) (N^D bisceope)))

(NP-NOM (NPR^N Dauid) (N^N cynincg))

(NP-NOM (NPR^N Seuerus) (N^N casere)

When the title is modified in any way, usually by a determiner, the title is labelled as an appositive.


(NP-NOM (NPR^N Chromatius)
	(NP-NOM-PRN (D^N se) (N^N heahgerefa)))

Temporal names

Note that the names of massdays and DOMES D+AG when written as two words are not treated as compounds.


(NP-ACC (NP-GEN (N|G hlafm+assan))
	(N^A d+ag))

(NP-ACC (NP-GEN (N^G domes))
	(N^A d+ag)))

Clausal syntax of Noun Phrases

Arguments vs. adjuncts
Subjects
Subjects in tensed clauses Subjects in non-tensed clauses Subjects in poetry Presentational THERE Expletive IT
Other arguments
Reflexives (-RFL) Foreign words and phrases
Dative of possession
Adjunct NPs
Temporal, locative and directional NPs Left-dislocated NPs Repeated subject pronouns

PPCME2 Table of contents

YCOE Table of contents

Arguments vs. adjuncts

NPs are divided into two main classes in YCOE, arguments and adjuncts. Adjuncts of most types have the extended label -ADT. Sub-classes of adjuncts that are more specifically labelled are temporals, locatives and directionals (-TMP, -LOC, -DIR) and left-dislocations (-LFD). Arguments are distinguished by the lack of an -ADT (or other adjunct) label; i.e., any NP not carrying an -ADT/-TMP/-DIR/-LOC/-LFD label is an argument. In general the only extended label arguments have is a case label (although if the head of the NP is not case-marked, not even this will be present), but the following finer distinctions among arguments are also made:

resumptive pronouns are labelled -RSP (e.g., NP-NOM-RSP) reflexives are labelled -RFL (e.g., NP-DAT-RFL) predicates are labelled -PRD (e.g., NP-NOM-PRD)

( (IP-MAT (CONJ ac)
          (NP-NOM (PRO^N he))             <--- case-marked argument NP
          (VBD bediglode)
          (ADVP (ADV swa) (ADV +teah))
          (NP-ACC (PRO$ his) (N^A d+ada)) <--- case-marked argument NP
          (NP-DAT (D^D +tam) (N^D casere) <--- case-marked argument NP
                  (NP-DAT-PRN (NPR^D Dioclitiane))
                  (CP-REL (WNP-NOM-1 (D^N se))
                          (C 0)
                          (IP-SUB (NP-NOM *T*-1)
                                  (BEDI w+as)
                                  (NP-NOM-PRD (NP-GEN (N^G deofles)) <--- case-marked predicate NP
                                              (N^N biggencga)))))
          (. .)) (ID AelfLives,+ALS_[Sebastian]:8.7))

( (CODE )
  (IP-MAT (ADVP-TMP (ADV^T D+aghwamlice))
          (NP-NOM (PRO^N he))
          (VBD gefylde)
          (NP (NP-GEN (PRO$ his) (NPR^G Drihtnes)) <--- non-case-marked argument NP
              (N +tenunge))
          (ADVP (ADV geornlice))
          (. ,)) (ID AelfLives,+ALS_[Sebastian]:8.6))

( (IP-MAT (CONJ and)
          (NP-NOM *con*)
          (PP (P on)
              (NP-DAT (NP-GEN (NPR^G Godes))
                      (N^D herungum)))
          (NP-ACC-RFL (PRO^A hi)             <--- reflexive argument NP
                      (ADJP-ACC (ADJ^A sylfe)))
          (VBDI gebysgodon)
          (. ,)) (ID AelfLives,+ALS_[Sebastian]:148.90))

( (CODE )
  (IP-MAT (NP-LFD (D +T+at) 
                  (CP-REL (WNP-ACC-1 (D^A +t+at))
                          (C 0)
                          (IP-SUB (NP-ACC *T*-1)
                                  (NP-NOM (PRO^N ic))
                                  (VBD gelyfde))))
          (NP-ACC-RSP (D^A +t+at))           <--- resumptive argument NP
          (NP-NOM (PRO^N ic))
          (VBP geseo)
          (. ,)) (ID AelfLives,+ALS[Agnes]:234.659))

(CP-THT-SPE-x (C +t+at)
	      (IP-SUB-SPE (NP-NOM (PRO^N +tu))
			  (VBD b+ade)
			  (NP-DAT-ADT (PRO$^D +tinum) (N^D bearnum))    <--- case-marked adjunct NP
			  (NP-GEN (N^G fyrstes))
			  (PP (P to) (D^I +ti)
			      (CP-ADV-SPE (C +t+at)
					  (IP-SUB-SPE (NP-NOM (PRO^N +tu))
						      (VBDI gelyfdest)
						      (NP-DAT (PRO$ heora) (ADJ^D leasum) (N^D gedwyldum)))))))

( (IP-MAT (CONJ ac)
          (NP-ADT (PRO us))                    <--- non-case-marked adjunct NP
          (VBDI bearn) 
          (NP-NOM (D^N +tis))
          (PP (P on)
              (NP-ACC (N^A mod)))
          (. .)) (ID AelfLives,+ALS[Peter's_Chair]:232.1208))

( (CODE )
  (IP-MAT (ADVP-TMP (ADV^T +Ta))
          (VBDI cw+a+d)
          (NP-NOM (NPR^N Sebastianus)
                  (, ,)
                  (NP-NOM-PRN (D^N se) (ADJ^N ges+aliga) (N^N martyr)))
          (, ,)
          (IP-MAT-SPE (PP (P Gif)
                          (CP-ADV-SPE (C 0)
                                  (IP-SUB-SPE (NP-NOM (PRO^N +tu))
                                          (NP-ACC (D^A +tisne) (N^A cr+aft))
                                          (VBPI healst))))
                      (NP-NOM (PRO^N +tu))
                      (BEPI bist)
                      (NP-DAT-RFL-ADT (PRO^D +te))        <--- reflexive adjunct NP
                      (ADJP-NOM (ADJ^N sylf))
                      (ADJP-NOM-PRD (ADJ^N unhal)))
          (. .)) (ID AelfLives,+ALS_[Sebastian]:265.159))

It is sometimes difficult to distinguish argument from adjunct NPs. This is especially true for dative pronouns where the distinction between an argument and a dative of interest is not always clear. The most difficult environment is in copular constructions (It was clear to me), and in this case we do not attempt to decide the issue but label all datives as arguments by default.

*difference*
Many of the functional labels used in the PPCME2 are not used in YCOE since case information is provided instead. In particular the class of adjuncts is less finely divided.

Subjects

Subjects in tensed clauses

The function SUBJECT is not explicitly labelled in most types of clauses. Because of the importance of distinguishing subjects from other arguments, however, we go to some trouble to ensure in the annotation of the prose texts that all tensed clauses except imperatives and UTON clauses have one and only one element that can be identified as the subject. In general this will be the nominative NP (NP-NOM) of which there will be only one (predicate nominative NPs are labelled as predicates). If no subject is present an empty subject is added. Subjects are not added in clauses with elision in which more than just the subject is elided. (The more complicated situation in poetic texts requires special rules for subjects).

Subjects in non-tensed clauses

Subjects in non-tensed clauses are treated as follows:

subjects are not indicated in infinitives unless the subject is overt or elided under conjunction; accusative subjects in the accusative and infinitive construction (whether case is marked or not) are also labelled -SBJ to distinguish them from complement accusatives
*difference*
note this is different from the PPCME2 where arbitrary PRO subjects are always indicated with ECM verbs

in small clauses and absolutes, non-nominative subjects are labelled -SBJ

in participle phrases only absolutes (PTP-case-ABS) have an overt subject and no empty subject is added in any circumstance

Subjects in poetry

Because of the frequent use of appositives and floating modifiers in poetry, and the substantive use of adjectives and other modifiers, deciding what the subject of a clause is not always straightforward. We use the following rules of thumb:

starting at the beginning of the clause, the first nominative nominal constituent not separated from the rest of the clause by punctuation (not counting punctuation around an intervening parenthetical or subordinate clause) is taken as the subject; all other nominal constituents are labelled NP-NOM-PRN and traced to the first one if the first nominative constituent is an ADJP, QP or NUMP and one or more nominative nominal constituents follow, separated from it by one or more constituents, all the nominative phrases in the clause are left at clause-level without -PRN label or tracing; the first nominative constituent is not labelled NP-NOM (and thereby specified as the subject) because it is not clear whether the ADJP/QP/NUMP is substantive or modifying in clauses without an NP-NOM but with a nominative adjective, quantifier or numeral (phrase) that could be interpreted as the subject, the latter is taken as the subject and labelled NP-NOM only if it is not co-referential with the subject of the previous clause; if it is co-referential, it is treated as a clause-level modifier and an empty *con* subject is added if the only nominative nominal constituent in the clause is separated from the rest of the clause by a comma (thereby reading as a possible right-dislocation) it is not clear whether this constituent is the subject or whether the subject is empty (elided or small pro); therefore, we label the nominal constituent NP-NOM without a -PRN label or tracing and also add an empty subject ((NP-NOM *con*) if the subject is coreferential with the subject of the previous clause and (NP-NOM *pro*) if it is not) in order to indicate both possible analyses

Presentational THERE

*difference*
Presentational THERE is not recognised in the YCOE, unlike in the PPCME2; +T+AR is always labelled ADVP-LOC and the accompanying nominative NP is taken as the subject of the clause.


( (CODE )
  (IP-MAT-SPE (ADVP-LOC (ADV^L +T+ar))
              (BEPI is)
              (NP-NOM (N^N wop) (CONJ and) (N^N wanung))
              (. FULL-STOP)) (ID AelfLives,+ALS_[Sebastian]:77.45))


( (CODE )
  (IP-MAT (ADVP-LOC (ADV^L +T+ar))
          (VBDI comon)
          (ADVP (ADV eac))
          (NP-NOM (PRO$ heora) (N^N magas))
          (. FULL-STOP)) (ID AelfLives,+ALS_[Sebastian]:40.25))

Nor is an expletive THERE ever added


(NODE (IP-MAT (BEPI Is)
              (NP-NOM (ADJ^N +a+dele) (N^N mynster))
              (PP (P in)
                  (NP (NPR Lindesse)))
              (. ;))
      (ID Bede3,Bede_3:9.182.15.359)) 


( (CODE )
  (IP-MAT (BEDI Wear+d)
          (ADVP-TMP (ADV^T +ta))
          (NP-NOM (Q^N mycel) (N^N angsumnyss))
          (PP (P on)
              (NP-DAT (Q^D eallum) (D^D +tam) (N^D folce)
                      (CP-REL (WNP-NOM-1 0)
                              (C +te)
                              (IP-SUB (NP-NOM *T*-1)
                                      (PP (P on)
                                          (NP-ACC (NPR^A God)))
                                      (VBDI gelyfdon)))))
          (, ,)
          (PP (P for)
              (NP-DAT (D^D +dam) (ADJ^D gramlicum) (N^D d+adum)))
          (. ;)) (ID AelfLives,+ALS_[Maccabees]:25.3025))


(NODE (IP-SUB (NP-NOM (Q^N sum) (N^N wer) <--- "there was a man who..."
                      (CP-REL *ICH*-1))
              (BEDS w+are)
              (CP-REL-1 (WNP-NOM-2 0)
                        (C +te)
                        (IP-SUB (NP-NOM *T*-2)
                                (NP-ACC (PRO$ his) (N^A wif))
                                (VBD fors+ade)
                                (, ,)
                                (PP (P swa)
                                    (CP-ADV (C +t+at)
                                            (IP-SUB (NP-NOM (PRO^N heo))
                                                    (MDD sceolde)
                                                    (NP-ACC (PRO^A hi))
                                                    (ADVP (ADV sceandlice))
                                                    (VB forlicgan)))))))
      (ID AelfLives,+ALS[Ash_Wed]:180.1574))

Expletive IT

PPCME2

Overt expletive HIT is not specially marked, except in combination with an extraposed subject clause or NP, when both the clause/NP and HIT have an index -x. Otherwise it is not distinguished from non-expletive subject HIT. Note that only HIT is treated as expletive. With a +T+AT subject in similar constructions the associated clause is is treated as an appositive.

*difference*
Expletive coindexing is done slightly differently from the PPCME2; expletive coindexing is indicated with -x rather than numbers.


(IP-MAT (NP-NOM it)
        (BEPI is)
        (VAG raining))

(NODE (IP-MAT (CONJ &)
              (NP-NOM-x (PRO^N hyt))
              (BEDI wear+d)
              (ADVP-TMP (ADV^T +ta))
              (NP-NOM-x (Q^N mycel) (N^N hungorgear))
              (PP (P on)
                  (NP-DAT (D^D +tam) (N^D lande)))
              (. ;))
      (ID Heptateuch,Gen:12.10.471))

( (IP-MAT (CONJ and)
          (NP-NOM-x (PRO^N hit))
          (VBDI gelamp)
          (CP-THT-x (C +t+at)
                    (IP-SUB (NP-NOM (D^N se) (N^N gedwola))
                            (VBDI rad)
                            (PP (P on)
                                (NP-DAT (D^D +d+are) (N^D wucan)))
                            (PP (P ymbe)
                                (NP-ACC (Q^A sum) (N^A +arende)))))
          (. ,))
      (ID AelfLives,+ALS[Ash_Wed]:50.1508))

See Expletive constructions for uses of expletives.

Other arguments

The function of arguments is generally not explicitly indicated in YCOE; rather case-information is provided when it is clear. Arguments themselves are indicated by a lack of the -ADT label which indicates an adjunct.

Reflexives (-RFL)

Any NP containing a non-possessive pronoun that is coreferential with the subject of its clause is labelled -RFL. These may be arguments or adjuncts. The -RFL label follows case, but precedes -ADT (for adjuncts).


( (IP-MAT (CONJ and)
          (NP-NOM *con*)
          (PP (P on)
              (NP-DAT (NP-GEN (NPR^G godes))
                      (N^D herungum)))
          (NP-ACC-RFL (PRO^A hi) 
                      (ADJP-ACC (ADJ^A sylfe)))
          (VBDI gebysgodon)
          (. ,)))


( (IP-MAT-SPE (NP-NOM (PRO^N Ic))
              (NP-RFL-ADT (PRO me))
              (VBP gebidde)
              (PP (P to)
                  (NP-DAT (D^D +dam) (NPR^D Gode)
                          (CP-REL (WNP-NOM-1 0)
                                  (C +te)
                                  (IP-SUB (NP-NOM *T*-1)
                                          (BEPI bi+d)
                                          (VAG eardigende)
                                          (PP (P on)
                                              (NP-DAT (N^D heofonum)))
                                          (PP (P mid)
                                              (NP-DAT (ADJ^D healicum) (N^D m+agen+trymme)))))))
              (. .)))

Foreign phrases acting as arguments

Latin names of peoples and places are POS tagged as NPR and so are simply labelled NPs in the usual way in the syntax. Latin non-names or names of things (Cathedra Sancit Petri, etc.), which are POS tagged FW, when NP arguments, are simply given the appropriate phrase label: NP-NOM for subjects, NP-PRD for predicates, and NP for other arguments. Direct speech is labelled QTP. Non-arguments at IP-level are labelled LATIN. Latin appositives are labelled as X phrases.


(IP-CON (NP-NOM *con*) 
	(NP-DAT (PRO^D him))
	(NP (NP (FW pater) (FW noster))         <--- argument
	    (CONJP (CONJ and)
		   (NP (N credon))))
	(VBPS t+ace))

(CP-THT (C +t+at)
	(IP-SUB (NP-NOM (Q^N +alc) (N^N preost))
		(HVPS habbe)
		(NP (NP (FW corporalem)         <--- argument
			(CONJP *ICH*-1))
		    (CONJP *ICH*-2))
		(PP (P +tonne)
		    (CP-ADV (C 0)
			    (IP-SUB (NP-NOM (PRO^N he))
				    (VBPS (VBP m+assige)))))
		(CONJP-1 (CONJ and) (FW subumlem))
		(PP (P under)
		    (NP (PRO$ his) (N alban)))
		(, ,)
		(CONJP-2 (CONJ and)
			 (NP-ACC (Q^A eal) (N^A m+assereaf)
				 (PTP-ACC (ADVP (ADV wur+dlice))
					  (VBN^A behworfen))))))
          (. .)) (ID elaw01,:33.36))

( (CODE )
  (IP-MAT (NP-NOM (FW Cathedra))                <--- subject
          (BEPI is)
          (VBN gereht) 
          (NP-NOM-PRD (N^N bisceopstol))
          (PP (P on)
              (NP-ACC (ADJ^A Englisc)))
          (. ,)) (ID AelfLives,+ALS[Peter's_Chair]:4.1052))

( (CODE )
  (IP-MAT (PP (P On)
              (NP-DAT (D^D +dysse) (N^D wucan)))
          (PP (P on)
              (NP-ACC (NPR^A Wodnesd+ag)))
          (, ,)
          (PP (ADVP (ADV swa))
              (P swa)
              (CP-CMPX (NP-NOM (PRO^N ge) 
                               (ADJP-NOM (ADJ^N sylfe)))
                       (VBPI witon)))
          (, ,)
          (BEPI is)
          (NP-NOM (FW caput) (FW ieiunii)
                  (, ,)
		  (CP-REL (WNP-NOM-1 (D^N +t+at))
			  (C 0)
			  (IP-SUB (NP-NOM *T*-1)
				  (BEPI is)
				  (PP (P on)
				      (NP-ACC (ADJ^A Englisc)))
				  (, ,)
				  (NP-NOM-PRD (N heafod) 
					      (NP-GEN (ADJ^G lenctenes) (N^G f+astenes))))))
          (. .)) (ID AelfLives,+ALS[Ash_Wed]:1.1488))

( (CODE )
  (IP-MAT (NP-NOM (NPR^N Anticristus))
          (BEPI is)
          (PP (P on)
              (NP (NPR L+aden)))
          (NP-PRD (FW contrarius) (FW Cristo)     <--- predicate
		  (, ,)
		  (CP-REL (WNP-NOM-1 (D^N +t+at))
			  (C 0)
			  (IP-SUB (NP-NOM *T*-1)
				  (BEPI is)
				  (PP (P on)
				      (NP-ACC (ADJ^A Englisc)))
				  (, ,)
				  (NP-NOM-PRD (NP-GEN (NPR^G Godes))
					      (N^N wi+dersaca)))))
          (. .)) (ID WHom,WHom_1b:7.4))

Dative of possession

Because of the difficulty of distinguishing the so-called "dative of possession" from datives of interest, we do not try to distinguish these two. All are labelled as adjuncts (NP-DAT-ADT) and put at IP-level.

There are two exceptions to this rule:

in copular constructions, all datives are treated as arguments datives of possession "trapped" by PP-conjunction are put in the specPP position; if both conjuncts have an associated dative, both are put in their respective specifier positions ( (IP-MAT-SPE (CONJ and) (NP-NOM (PRO^N +tu)) (VBDI lutodest) (PP (P o+d) (NP-ACC (D^A +tis))) (PP (P on) (NP-DAT (D^D +tam) (ADJ^D la+dum) (N^D Cristendome))) (, ,) (PP (NP-DAT (D^D +tam) (N^D godum)) (PP (P to) (NP (N teonan))) (CONJP (CONJ and) (PP (NP-DAT (PRO^D me)) (P to) (NP (N un+tearfe))))) (. .)) (ID AelfLives,+ALS_[Sebastian]:410.248))

Note that most dative NPs in specPP position are complements of the PP (HIM TO, etc.); see NPs in specPP.

Adjunct NPs

NPs which are not arguments are labelled -ADT or with a more specific adjunct label (see Temporal, locative or directional NPs and Left-dislocations) if appropriate. Case if marked comes before the -ADT label.


( (CODE )
  (IP-MAT (NP-ACC (NPR^A Sebastianum))
          (NP-NOM (PRO^N he))
          (VBD gesette)
          (NP-DAT-ADT (PRO^D him) (Q^D eallum))
          (PP (P to)
              (NP (N mundboran)))
          (. .)))

( (CODE )
  (IP-MAT (INTJ Hw+at)
          (ADVP-TMP (ADV^T +da))
          (NP-NOM (NPR^N Tiburtius))
          (ADVP (ADV bealdlice))
          (VBD eode)
          (PP (P ofer)
              (NP-ACC (D^A +da) (VAG^A byrnendan) (N^A gleda)))
          (, ,)
          (NP-DAT-ADT (ADJ^D unforb+arnedum) (N^D fotum))
          (. ,)))

( (IP-MAT (CONJ and)
          (NP-NOM *con*)
          (PP (ADVP-LOC (ADV^L +t+ar))
              (P binnan))
          (VBD wunode)
          (NP-TMP (NUMP (NUM fif) (CONJ and) (NUM twentig))
                  (NP-GEN (N^G geara)))
          (. ,)))

Temporal, locative and directional NPs

Adjunct NPs with temporal, locative and directional meaning are labelled with -TMP, -LOC, -DIR in the same way as ADVPs. Case, if indicated, precedes the adjunct label. In fact, only temporals occur with any regularity (so far).


( (IP-MAT (CONJ and)
          (NP-NOM *con*)
          (VBD wacode)
          (NP-ACC-TMP (Q^A ealle) (D^A +ta) (N^A niht))
          (PP (P ofer)
              (NP-ACC (NP-GEN (D^G +d+as) (N^G wodan))
                      (N^A lic)))
          (. ,)) (ID AelfLives,+ALS_[Maur]:206.402))

( (IP-MAT (CONJ and)
          (NP-NOM *con*)
          (PP (ADVP-LOC (ADV^L +t+ar))
              (P binnan))
          (VBD wunode)
          (NP-TMP (NUMP (NUM fif) (CONJ and) (NUM twentig))
                  (NP-GEN (N^G geara)))
          (. ,)) (ID AelfLives,+ALS[Peter's_Chair]:7.1037))

( (CODE )
  (IP-MAT (ADVP-TMP (ADV^T +Ta))
          (VBDI gerad)
          (NP-NOM (NPR^N +A+delwald)
                  (NP-NOM-PRN (NP-GEN (PRO$ his) (N^G f+adran))
                              (N^N sunu)))
          (, .)
          (NP-ACC-DIR (D^A +tone) (N^A ham) 
                      (PP (PP (P +at)
                              (NP (NPR Winburnan)))
                          (, ,)
                          (CONJP (CONJ &)
                                 (PP (P +at)
                                     (NP-DAT (NPR Tweoxn) (NPR^D eam)))))) 
          (PP (P butan)
              (NP (NP-GEN (NP-GEN (D^G +d+as) (N^G cyninges))
                          (CONJP *ICH*-1))
                  (N leafe)
                  (CONJP-1 (CONJ &)
                           (NP-GEN (PRO$ his) (N^G witena)))))
          (. .)) (ID ChronA,:901.6.1160))

( (CODE )
  (IP-MAT (NP-NOM-x *exp*) 
          (NP-LOC (NP-GEN (D^G +Tisses) (ADJ^G ylcan) (NPR^G Bonefacies)) 
                  (N cyrican))
          (VBDI gelamp)
          (, ,)
          (CP-THT-x (C +t+at)
		    (IP-SUB ...))))

Left-dislocated NPs

Left-dislocated NPs are labelled -LFD. The associated resumptive element (usually a pronoun) is labelled -RSP. Both labels apply after case, if present. Special case-marking rules apply for left-dislocations.


( (CODE )
  (IP-MAT (NP-NOM-LFD (D^N +Da)
                      (CP-REL (WNP-NOM-1 0)
                              (C +de)
                              (IP-SUB (NP-NOM *T*-1)
                                      (BEDI w+aron)
                                      (RP+VBN^N for+dferede)
                                      (PP (PP (P for)
                                              (NP-DAT (NUM hund) (N^D gearum)))
                                          (CONJP (CONJ o+d+don)
                                                 (ADVP-TMP (ADV^T gyt) (ADV^T firnor)))))))
          (, ,)
          (ADVP (ADV wel))
          (NP-NOM-RSP (D^N +ta))
          (MDPI magan)
          (BE beon)
          (ADVP-TMP (ADV^T nu))
          (VBN^N gecl+ansode)
          (. .)) (ID WHom,WHom_4:29.120))

( (IP-MAT (CONJ &)
          (NP-NOM-LFD (D^N se)
                      (CP-REL (WNP-NOM-1 0)
                              (C +de)
                              (IP-SUB (IP-SUB (NP-NOM *T*-1)
                                              (NP-ACC (N^A woh))
                                              (VBPI drif+d))
                                      (CONJP (CONJ &)
                                             (IP-CON (NP-NOM *T*-1)
                                                     (VB geswican)
                                                     (NEG+MDP nele))))))
          (, ,)
          (NP-NOM-RSP (PRO^N he))
          (MDPI sceal)
          (HV habban)
          (NP-GEN-ADT (D^G +d+as)) 
          (NP-ACC (ADJ^A ece) (N^A wite))
          (. .)) (ID WHom,WHom_6:17.254))

Repeated subject pronouns

When a subject pronoun is repeated within the same clause, the original pronoun is labelled NP-NOM and the second as NP-NOM-RSP.


(NODE (IP-SUB-SPE (NP-NOM (PRO^N +du))     <--- first pronoun
                  (NP-ACC (NP-ACC (D^A +tone) (N^A mangengan))
                          (CONJP (CONJ &)
                                 (NP-ACC (D^A +tone) (N^A wi+tfeohtend)))
                          (CONJP (CONJ &)
                                 (NP-ACC (D^A +tone)
                                         (N^A forhycgend)
                                         (NP-GEN (PRO$^G ura) (N^G goda)))))
                  (NP-NOM-RSP (PRO^N +du)) <--- repeated pronoun
                  (NP (PRO me))
                  (VB helan)
                  (MDDI woldest)
                  (, ,)
                  (ADVP (ADV swy+dor)
                        (PP (P +tonne)
                            (CP-CMP-SPE (WADVP-1 0)
                                        (C 0)
                                        (IP-SUB-SPE=0 (ADVP *T*-1)
                                                      (NP-DAT (PRO$^D minum) (N^D +degnum))
                                                      (VB secgean))))))
      (ID Bede1,Bede_1:7.36.3.115)) 

(NODE (IP-MAT (CONJ Ac)
              (NP-NOM (PRO^N ic))   <--- first pronoun
              (, ,)
              (PP (ADVP (ADV swa))
                  (P swa)
                  (CP-CMPX (NP-NOM (ADJ^N so+dsagal) (N^N st+arwritere))))
              (, ,)
              (NP-ACC (D^A +ta) (N^A +ting)
                      (, ,)
                      (CP-REL (WNP-NOM-2 0)
                              (C +te)
                              (IP-SUB (NP-NOM *T*-2)
                                      (PP (PP (P be)
                                              (NP-DAT (PRO^D him)))
                                          (CONJP (CONJ o+d+de)
                                                 (PP (P +turh)
                                                     (NP-ACC (PRO^A hine)))))
                                      (BEN^N gewordene)
                                      (BEDI w+aron))))
              (, ,)
              (NP-NOM-RSP (PRO^N ic)) <--- repeated pronoun
              (VBDI awrat)
              (. ,))
      (ID Bede3,Bede_3:14.206.5.632))

Prepositional phrases (PP)

Particle plus preposition sequences
Multi-word prepositions (MID +TY, BE SU+TAN, TO MIDDES, etc.)
Prepositions with empty objects
+TY L+AS +TE
Adverbial clauses headed by SWA
SWA +T+AT purpose/result clauses Temporal SWA clauses including SONA SWA
Elements in specPP NPs in specPP ADVPs in specPP
Adverbial clauses
PPCME2 Table of contents
YCOE Table of contents

As in the PPCME2 prepositions take both NP and clausal complements. Most adverbial subordinate clauses in the PPCME2/YCOE are headed by prepositions. See Adverbial clauses for other types.


( (IP-MAT (CONJ and)
          (NP-NOM *con*)
          (PP (P mid)
              (NP (N lufe)))
          (VBDI wur+dodon)
          (, ,)
          (PP (P for+don)
              (CP-ADV (C +te)
                      (IP-SUB (NP-NOM (NPR^N God))
                              (NP-ACC (PRO^A hine))
                              (VBD lufode))))
          (. .)) (ID AelfLives,+ALS_[Sebastian]:15.13))

(NODE (IP-MAT-SPE (PP (P Gif)
		      (CP-ADV-SPE (C 0)
                                  (IP-SUB-SPE (NP-NOM (PRO^N +tu))
					      (NP-ACC (D^A +tisne) (N^A cr+aft))
					      (VBPI healst))))
		  (NP-NOM (PRO^N +tu))
		  (BEPI bist)
		  (NP-DAT-RFL-ADT (PRO^D +te))
		  (ADJP-NOM (ADJ^N sylf))
		  (ADJP-NOM-PRD (ADJ^N unhal)))
      (. .)) (ID AelfLives,+ALS_[Sebastian]:265.159))

A preposition which does not take NP complements dominates a clause no matter how much material has been elided from the clause. Equal-sign coindexing is used, as usual with elision constructions to indicate that the clause is not complete.


( (IP-MAT-0 (CONJ and)
            (NEG+BEPI nis)
            (NP-NOM (NEG+Q^N nan) (ADJ^N o+der) (NPR^N God))
            (PP (P buton)
                (CP-ADV (C 0)
                        (IP-SUB=0 (NP-NOM (PRO^N he) (FP ana)))))
            (. .))
      (ID AelfLives,+ALS_[Apollinaris]:116.2753)) 


( (IP-MAT-0 (ADVP-TMP (ADV^T +ta))
            (NEG+HVDI n+afdon)
            (NP-NOM (PRO^N hi))
            (NP-ACC (NEG+Q^A nan) (N^A win))
            (PP (P buton)
                (CP-ADV (C 0)
                        (IP-SUB=0 (PP (P on)
                                      (NP-DAT (NUM^D anum) (VBN^D gewealdenan) (N^D butruce))))))
            (. .))
      (ID AelfLives,+ALS_[Maur]:273.443))

Particle plus preposition sequences

All sequences of two prepositions for which the first preposition also functions as an adverbial particle are analyzed as a particle-preposition sequence unless the particle clearly doesn't go with the PP (e.g., they walked around in the park). The particle is included in the dominating PP in specPP position.


(PP (RP up) (P to)
    (NP ...))

( (CODE )
  (IP-MAT (NP-NOM (PRO^N He))
          (VBD eode)
          (ADVP-TMP (ADV^T +ta))
          (ADVP-TMP (ADV^T eft))
          (PP (RP ut) (P to)
              (NP-DAT (D^D +tam) (ADJ^D earmum) (N^D wodum)))
          (. ,)) (ID AelfLives,+ALS_[Maur]:203.403))

( (IP-MAT (CONJ and)
          (NP-NOM (PRO^N heo))
          (VBDI hnah)
          (PP (RP adune) (P to)
              (NP-DAT (NP-GEN (NPR^G Sebastianes))
                      (N^D fotum)))
          (PP (P mid)
              (NP-DAT (ADJ^D fullum) (N^D geleafan)))
          (. .)) (ID AelfLives,+ALS_[Sebastian]:90.58))

Sequences for which the first preposition does not also function as a particle are treated as multi-word_prepositions.

Multi-word prepositions (MID +TY, BE SU+TAN, TO MIDDES, TO...WEARD, etc.)

Most potential sequences of preposition-preposition are treated as particle + preposition sequences (e.g., UP ON, IN TO, etc.).

But when the first preposition is not on the adverbial particles list (e.g., FORAN TO), then the sequence is treated as a preposition taking a prepositional phrase complement.


(PP (P foran)
    (PP (P to)
	(NP ...)))

Note that uniquely, WEARD, when not fused to its co-preposition, follows the nominal complement. Its structure is as follows. (See also BE...TWEONUM).


(PP (P to)
    (PP (NP (NPR Babilonian))
	(P werd)))

(PP (P wi+d)
    (PP (NP-GEN (D^G +t+as) (N^G fyres))
	(P weard)))

When the second element is not a preposition, it is POS tagged literally according to its part-of-speech, and both it and the preposition are immediately dominated by the same PP node in the parsing. The two most common types are preposition-determiner (MID +TY, FOR +TI, etc.), and preposition-adverb (ON UFAN, BE SU+TAN, etc.), but the more unusual TO MIDDES and ON MIDDAN also fall into this class. These sequences are treated the same way, whether they take an NP or a clausal complement. In general, P+D takes a clause, while P+ADV takes an NP. When these sequences are written as a single word, they are labelled as prepositions if they take a complement and adverbs otherwise.


(PP (P prep) (D determiner)
    (NP/clause ...))

(PP (P prep) (ADV adverb)
    (NP ...))

(PP (P to) (N^G middes)
    (NP ...))

(PP (P mid) (D^I +ty)
    (CP-ADV ...))

(PP (P be) (ADV su+tan)
    (NP-DAT (N^D ...)))

(PP (P wi+t) (ADV nor+tan)
    (NP-ACC (N^A ...)))

(PP (P on) (ADJ middan)
    (NP ...))


( (IP-MAT (CONJ ac)
          (NP-NOM (PRO^N hi))
          (BEDI wurdon)
          (VBN^N geh+alede)
          (PP (P fram)
              (NP (PRO$ heora) (N untrumnysse)))
          (PP (P mid) (D^D +tam)
              (CP-ADV (C +te)
                      (IP-SUB (NP-NOM (D^N se) (N^N m+assepreost))
                              (NP-ACC (PRO^A hi))
                              (PP (P mid)
                                  (NP-DAT (D^D +tam) (N^D fulluhte)))
                              (VBDI a+twoh))))
          (. .))
      (ID AelfLives,+ALS_[Sebastian]:144.88)) 

(NODE (IP-SUB (NP *T*-1)
              (NP-NOM (NPR^N Thabitas))
              (NP-DAT-ADT (PRO^D him))
              (VBD worhte)
              (PP (P +ar) (D^I +tan)
                  (CP-ADV (C +te)
                          (IP-SUB (NP-NOM (PRO^N heo))
                                  (VBDS gewite)))))
      (ID AelfLives,+ALS[Peter's_Chair]:63.1095)) 

(NODE (IP-MAT (ADVP-TMP (ADV^T +Tonne))
              (PP (P be)
                  (ADV nor+tan)
                  (NP (NPR Syria)))
              (BEPI sindon)
              (NP-NOM (D^N +ta)
                      (N^N beorgas)
                      (CP-REL (WNP-1 0)
                              (C +te)
                              (IP-SUB (NP-NOM (MAN^N mon))
                                      (IP-SMC (NP-SBJ *T*-1)
                                              (NP-PRD (NPR Tauros)))
                                      (VBPI h+att)))))
      (ID Orosius,:1.10.20.42))

Note that a clausal complement headed by +TE can be extraposed in which case it is traced to the PP.


(IP (...)
    (PP (P for) (D^D +tam)
	(CP-ADV *ICH*-1))
    (...)
    (CP-ADV-1 (C +te)
	      (IP-SUB ...)))


(NODE (IP-MAT (NP-NOM (D^N +Da) (ADJ^N manigfealdan) (N^N w+astmas))
              (BEDI w+aron)
              (PP (P for) (D^D +tam)
                  (CP-ADV *ICH*-1))
              (ADJP-NOM-PRD (ADJ^N swi+tost))
              (CP-ADV-1 (C +de)
                        (IP-SUB (NP-NOM (NPR^N Iordanis)
                                        (NP-NOM-PRN (D^N seo) (N^N ea)))
                                (NP-INS-TMP (Q^I +alce) (N^I geare))
                                (NP-ACC (D^A +t+at) (N^A land) (ADJ^A middeweard))
                                (RP+VBDI oferfleow)
                                (PP (P mid)
                                    (NP-INS (NP-GEN (N^G fotes))
                                            (ADJ^I +ticce)  (N^I flode)))))
              (. ,))
      (ID Orosius,:3.22.32.348))

Prepositions with empty objects

When a preposition/particle has no overt complement, but the meaning seems to require a complement, and one is implicit in the context (either from the previous token, or another clause in the same token), it is not clear whether these should be taken as prepositions or particles. Thus they are labelled RPX, and no complement is indicated.


( (CODE )
  (IP-MAT (ADVP-TMP (ADV^T +Ta))
          (NP-NOM (D^N se) (ADJ^N halga) (N^N wer))
          (VBDI nam)
          (NP-ACC (D^A +t+at) (N^A hors))) (ID GDC1,:10.78.24.817))
( (IP-MAT (CONJ &)
          (NP-NOM *con*)
          (NP-ACC (ADJ^A healfne) (D^A +tone) (N^A weg))
          (RPX on)        <--- ON +T+AT HORS
          (VBD geferde)
          (. .)) (ID GDC1,:10.78.24.818))


( (CODE )
  (IP-MAT (ADVP-TMP (ADV^T +Ta))
          (RP+AXDI ongunnon)
          (NP-NOM (NP-NOM (D^N +ta) (N^N nedlingas))
                  (CONJP (CONJ &)
                         (NP-NOM (D^N +ta) (N^N scipmen))))
          (NP-ACC (D^A +ta) (N^A oncras))
          (RP upp)
          (VB teon)
          (. ,)) (ID Bede3,Bede_3:13.200.11.575))
(IP-MAT (CONJ &)
        (NP-NOM *con*)
        (PP (P in)
            (NP-ACC (D^A +tone) (N^A s+a)))
        (VBDI syndon) 
        (. ;))
( (IP-MAT (NP-NOM *con*) 
          (MDDI woldon)
          (NP-ACC (D^A +t+at) (N^A scip))
          (RPX mid)           <--- MID +TA ONCRAS
          (VB gef+astnian)
          (. .)) (ID Bede3,Bede_3:13.200.11.577))

( (IP-MAT (NP-GEN-1 (D^G +tara))
          (BEDI w+aron)
          (NP-NOM (NP-GEN *ICH*-1)
                  (NUM syx))
          (NP-NOM-PRD (N^N st+alhranas)
                      (, ,)
                      (CP-REL (WNP-NOM-2 (D^N +da))
                              (C 0)
                              (IP-SUB (NP-NOM *T*-2)
                                      (BEPI beo+d)
                                      (ADJP-NOM-PRD (ADV swy+de) (ADJ^N dyre))
                                      (PP (P mid)
                                          (NP-DAT (NPR^D Finnum)))
                                      (, ,)
                                      (PP (P for) (D^D +d+am)
                                          (CP-ADV (C 0) 
                                                  (IP-SUB (NP-NOM (PRO^N hy))
                                                          (VBPI fo+d)
                                                          (NP-ACC (D^A +ta) (ADJ^A wildan) (N^A hranas))
                                                          (RPX mid))))))))  <--- MID ST+ALHRANAS 
          (. .)) (ID Orosius,:1.15.9.156))

+TY L+AS +TE

Clauses headed by +TY L+AS or +TE L+AS unless are treated in a similar way to the preposition plus determiner clauses in that +TY/+TE is literally labelled as a determiner and immediately dominated by the PP node. Again this is not meant to represent a linguistic analysis; it is simply a notational device.


(PP (D^I +ty) (P l+as)
    (CP-ADV (C +te)
            (IP-SUB ...)))

Adverbial clauses headed by SWA

SWA +T+AT purpose/result clauses


( (CODE )
  (IP-MAT (ADVP-TMP (ADV^T +Ta))
          (VBDI feol)
          (NP-NOM (Q^N sum) (N^N preost))
          (ADVP (ADV f+arlice))
          (PP (P of)
              (NP-DAT (D^D +tam) (N^D weorce)))
          (, ,)
          (PP (P swa)
              (CP-ADV (C +t+at)
                      (IP-SUB (NP-NOM (PRO^N he))
                              (ADJP-NOM (ADJ^N samcucu))
                              (VBDI l+ag)
                              (, ,)
                              (ADJP-NOM (NP-DAT (VAG^D sweltendum))
                                        (ADJ^N gelic)))))
          (. ,)) (ID AelfLives,+ALS_[Maur]:163.375))

Temporal SWA clauses including SONA SWA

Bare SWA clauses can be temporal although these are not as common as temporal AS clauses later become. Most temporal SWA clauses are headed by SONA SWA.


( (IP-MAT (CONJ &)
          (PP (P swa)
              (CP-ADV (C 0)
                      (IP-SUB (NP-NOM (D^N +tis))
                              (VBN gedon)
                              (BEPI by+t))))
          (, ,)
          (NP-NOM (PRO^N ic))
          (VBP ga)
          (PP (P +after)
              (NP (PRO +te)))
          (. .))
      (ID GDC1,:4.36.30.342)) 

(NODE (IP-MAT (ADVP-TMP (ADV^T +Ta))
              (ADVP (ADV semninga))
              (NP-NOM-LFD (D^N se) (ADJ^N ylca) (NPR^N Iulianus))
              (, ,)
              (PP (P swa)
                  (CP-ADV (C 0)
                          (IP-SUB (NP-NOM (PRO^N he))
                                  (VBDI geseah)
                                  (NP-ACC (D^A +tone)
                                          (NP-GEN (NPR^G Godes))
                                          (N^A +teowan)))))
              (, ,)
              (NP-NOM-RSP (PRO^N he))
              (VBDI forseah)
              (NP-ACC (PRO^A hine))
              (ADVP-TMP (ADV^T sona))
              (PP (P for)
                  (NP (PRO$ his) (N gegerelan)))
              (. FULL-STOP))
      (ID GDC1,:4.37.17.353)) 

( (IP-MAT (CONJ ac)
          (PP (ADVP-TMP (ADV^T sona))
              (P swa)
              (CP-ADV (C 0)
                      (IP-SUB (NP-NOM (PRO^N he))
                              (VBD andette)
                              (PP (P mid)
                                  (NP-DAT (Q^D ealre) (N^D heortan)))
                              (NP-ACC (D^A +ta) (ADJ^A halgan) (N^A +trynnysse))
                              (PP (P on)
                                  (NP-DAT (D^D +tam) (N^D fulluhtba+de))))))
          (, ,)
          (ADVP-TMP (ADV^T +ta))
          (BEDI wear+d)
          (NP-NOM (PRO^N he))
          (VBN geh+aled)
          (PP (P fram)
              (NP-DAT (Q^D eallum) (PRO$ his) (N^D sarnyssum)))
          (. ,)) (ID AelfLives,+ALS_[Sebastian]:135.84))

Elements in specPP

There are a number of cases where an element can appear before the head in a PP. Commonly this is an modifying adverb or particle but it can also be an NP, either the complement of the preposition or a modifying element. The elements taken to occupy specPP position are shown as sisters of the head. Apart from particles which never project a phrase, even single word modifiers in these cases are given phrase labels (unlike in other phrases (NP/QP/NUMP/ADJP/ADVP).


(PP (RP particle)
    (P prep)
    (NP complement))

(PP (ADVP (ADV adverb))
    (P prep)
    (NP complement))

(PP (NP-DAT (N^D pronoun))
    (P prep))

(PP (NP-DAT-EXT (N^D noun))
    (P prep)
    (NP complement))

NPs in specPP

There are a number of constructions in which an NP appears before the head preposition in a PP. In the first type the complement of the preposition appears before it. In most cases this is a pronoun.


(NODE (IP-SUB (NP-NOM (PRO^N he))
              (PP (NP-DAT (PRO^D him))
                  (P to))
              (VBDS come))
      (ID AelfLives,+ALS_[Sebastian]:152.92)) 

(NODE (IP-MAT (ADVP-TMP (ADV^T +Ta))
              (VBDI sang)
              (NP-NOM (PRO^N he))
              (PP (NP-DAT (PRO^D him))
                  (P ofer))
              (NP (NP (FW Pater) (FW Noster))
                  (CONJP (CONJ and)
                         (NP (N Credan))))
              (. ,))
      (ID AelfLives,+ALS_[Sebastian]:359.217))

In the other cases, this NP is not the complement of the preposition, which occurs its usual position following the preposition. The two types can be distinguished by the fact that in the first type no NP follows the preposition, while in the second there is an NP on both sides of the preposition.

with some cases of datives of possession (NODE (IP-SUB (NP-NOM (PRO^N he)) (NP-ACC (NPR^A Clementem)) (PP (PP (NP-DAT-RFL (PRO^D him)) <--- dative of possession (P to) (NP-DAT (N^D fultome) (NP (ADJ $godcundre) (N lare)))) (CONJP (CONJ &) (PP (NP-DAT-RFL (PRO^D him)) <--- dative of possession (P to) (NP-DAT (N^D +afterfylgende))))) (VBD gehalgade)) (ID Bede2a,Bede_2:4.106.20.144)) in Time expressions with +T+AS ( (IP-MAT (CONJ and) (NP-NOM *con*) (VBDI het) (PP (NP-GEN (D^G +t+as)) <--- +t+as (P on) (NP-ACC (N^A mergen))) (IP-INF (VB m+assian) (PP (P for) (NP (PRO$ his) (N sawle)))) (. ,)) (ID AelfLives,+ALS_[Maur]:206.406)) with NPs of extent (NODE (IP-MAT (ADVP-LOC (ADV^L Her)) (NP-NOM (N^N sunne)) (VBD a+tiestrode) (PP (NP-DAT-EXT (NUM xiiii) (N^D dagum)) (P +ar) (NP (FW Kalend+a) (FW Martii))) (PP (P from) (NP-DAT (N^D +armergenne))) (PP (P o+t) (NP-ACC (N^A undern))) (. .)) (ID ChronA,:538.1.186))

ADVPs in specPP

Adverbs preceding a preposition are usually modifying, but can be complements in the type +T+AR FRAM, HER ABOUE, etc. The first element in the sequences +TA +TA, SWA SWA, and SONA SWA is in specPP. When these are written as single words, however, they are simply tagged as prepositions.


(NODE (IP-CON (NP-NOM *con*)
              (VBDI geceas)
              (NP-ACC (D^A +ta) (ADJ^A cl+anan))
              (PP (P to)
                  (NP-DAT (PRO$ his) (ADJ^D cl+anum) (N^D +teowdome)))
              (, ,)
              (PP (ADVP (NEG+ADV na))
                  (PP (P to)
                      (NP (NP-GEN (N^G nytena))
                          (N offrunge)))
                  (CONJP (CONJ ac)
                         (PP (P to)
                             (NP-DAT (PRO$ his) (ADJ^D agenum) (N^D lichaman))))))
      (ID AelfLives,+ALS[Peter's_Chair]:224.1202))

(NODE (IP-SUB (NP-NOM (PRO^N ge))
              (BEPS beon)
              (VBN^N gescrifene)
              (PP (PP (P on)
                      (NP-DAT (D^D +dissere) (N^D wucan)))
                  (, ,)
                  (CONJP (CONJ o+d+de)
                         (PP (ADVP (ADV huru))
                             (P on)
                             (NP-DAT (D^D +d+are) (ADJ^D o+dre))))))
      (ID AelfLives,+ALS[Ash_Wed]:289.1641))

(NODE (IP-SUB (NP-NOM (PRO^N he))
              (VB faran)
              (MDD sceolde)
              (PP (ADVP-EXT (ADV feor))
                  (P fram)
                  (NP-DAT (D^D +d+are) (N^D byrig))))
      (ID AelfLives,+ALS_[Sebastian]:435.267))

( (IP-MAT (CONJ ac)
          (PP (ADVP-TMP (ADV^T sona))   <--- SONA SWA
              (P swa)
              (CP-ADV (C 0)
                      (IP-SUB (NP-NOM (PRO^N he))
                              (VBD andette)
                              (PP (P mid)
                                  (NP-DAT (Q^D ealre) (N^D heortan)))
                              (NP-ACC (D^A +ta) (ADJ^A halgan) (N^A +trynnysse))
                              (PP (P on)
                                  (NP-DAT (D^D +tam) (N^D fulluhtba+de))))))
          (, ,)
          (ADVP-TMP (ADV^T +ta))
          (BEDI wear+d)
          (NP-NOM (PRO^N he))
          (VBN geh+aled)
          (PP (P fram)
              (NP-DAT (Q^D eallum) (PRO$ his) (N^D sarnyssum)))
          (. ,)) (ID AelfLives,+ALS_[Sebastian]:135.84))

( (CODE )
  (IP-MAT (INTJP (WPRO Hw+at))
          (ADVP-TMP (ADV^T +da))
          (NP-NOM (NPR^N Nicostratus))
          (BEDI wear+d)
          (ADVP (ADV swi+de))
          (VBN afyrht)
          (, ,)
          (PP (ADVP-TMP (ADV^T +ta))      <--- +TA +TA
              (P +da)
              (CP-ADV (C 0)
                      (IP-SUB (NP-NOM (PRO^N he))
                              (NP-ACC-1 (D^A +t+at) (N^A wundor))
                              (VBDI geseah)
                              (IP-SMC (NP-ACC-SBJ *ICH*-1)
                                      (PP (P on)
                                          (NP-DAT (PRO$ his) (N^D wife)))
                                      (VBN gedon)))))
          (. ,)) (ID AelfLives,+ALS_[Sebastian]:112.68))

( (IP-MAT-SPE (CONJ and)
              (NP-NOM *con*)
              (VBDI nam)
              (NP-ACC (PRO$ his) (ADJ^A agene) (N^A swystor))
              (, ,)
              (PP (P to)
                  (NP-DAT (PRO$ his) (ADJ^D fulum) (N^D synscype)))
              (, ,)
              (PP (ADVP (ADV swa))        <--- SWA SWA
                  (P swa)
                  (CP-CMPX-SPE (NP-NOM (PRO^N ge))
                           (VBPI r+ada+d)
                           (PP (P on)
                               (NP-DAT (PRO$^D eowrum) (N^D gerecednyssum)))))
              (. .)) (ID AelfLives,+ALS_[Sebastian]:172.102))

(NODE (IP-SUB (NP *T*-2)
              (NP-NOM (D^N +teos) (N^N boc))
              (VBPI seg+d)
              (ADVP (ADV swutelice))
              (PP (ADVP-LOC (ADV^L her))   <--- adverbial complement
                  (P b+aftan)))
      (ID AelfLives,+ALS_[Maur]:131.357))

( (IP-MAT (CONJ and)
          (NP-NOM (PRO^N he))
          (VBDI s+at)
          (PP (ADVP-LOC (ADV^L +t+ar))     <--- adverbial complement
              (P onuppan))
          (. ,))
  (ID AelfLives,+ALS[Pr_Moses]:23.1664))

Quotation Phrases (QTP)

Direct speech is indicated in two ways in the YCOE. IPs and CPs have an extended label -SPE added; but other sequences, either a phrase alone or any group of constituents, is labelled QTP. QTPs are generally the direct complement of a verb of saying, but like direct speech IPs and CPs, they may also appear as their own token in long sequences of speech. Complements of verbs of saying which consist of or include foreign words or phrases are labelled QTP.


(NODE (IP-MAT (NP-NOM (NPR^N Agathes))
              (VBDI cw+a+d)
              (, ,)
              (QTP (NP-NOM (NPR^N Crist)          <--- single phrase QTP
                           (NP-NOM-PRN (NP-GEN (NPR^G Godes))
                                       (N^N sunu))))
              (. .))
      (ID AelfLives,+ALS[Agatha]:163.904)) 

(NODE (IP-MAT (ADVP-TMP (ADV^T +Ta))
              (VBDI cw+adon)
              (NP-NOM (PRO^N hi))
              (QTP (INTJ amen))
              (. ,))
      (ID AelfLives,+ALS[Forty_Soldiers]:255.1427)) 

(NODE (IP-MAT (NP-NOM (PRO^N Heo))
              (VBDI cw+a+d)
              (ADVP-TMP (ADV^T +ta))
              (PP (P to)
                  (NP-DAT (D^D +dam) (N^D cnihte)))
              (, ,)
              (QTP (INTJ eala)                     <--- more than one phrase
                   (NP-NOM-VOC (PRO^N +tu)
                               (ADJ^N forcu+dost)
                               (NP-GEN (N^G manna))))
              (. .))
      (ID AelfLives,+ALS[Ash_Wed]:197.1588)) 

(NODE (IP-MAT (CONJ And)
              (NP-DAT (PRO^D him))
              (VBD andwyrde)
              (NP-NOM (D^N +t+at) (N^N heafod))
              (, ,)
              (QTP (ADVP-LOC (ADV^L Her))
                   (, ,)
                   (ADVP-LOC (ADV^L her))
                   (, ,)
                   (ADVP-LOC (ADV^L her)))
              (. ;))
      (ID AelfLives,+ALS_[Edmund]:150.3669)) 

( (CODE )
  (CP-QUE (WNP-ACC-1 (WADJ^A Hwilc) (N^A fulluht))
          (IP-SUB (NP-ACC *T*-1)
                  (VBD sealde)
                  (NP-NOM (PRO^N he)))
          (. ?)) (ID CathHomII,+ACHom_II,_3:25.208.625))
( (QTP (NP-NOM (PRO$ His) (ADJ^N agen) (N^N fulluht))
       (, .)
       (PP (PP (P on)
               (NP-DAT (NP-GEN (NEG+Q^D nanre) (N^G synne)) 
                       (N^D forgifenysse)))
           (, .)
           (CONJP (CONJ ac)
                  (PP (P to)
                      (NP (N behreowsunge) (, .) (CONJ and) (N gearcunge)
                          (, .)
                          (PP (P to)
                              (NP-GEN (NPR^G Cristes)
                                      (NP (N fulluhte))))))))
       (. ;)) (ID CathHomII,+ACHom_II,_3:25.208.626))

X Phrases (XP)

The XP label is a device to allow a mismatch between the grammatical category of a phrase and its use in the clause. It is used in the following cases:

a non-nominal fills a normally nominal argument position, usually as a predicate in a naming construction; in this case, the XP also has a -PRD label; note that foreign language sequences used as arguments are not dominated by XP, but rather directly by NP. ( (IP-MAT (CONJ and) (NP-NOM (D^N se) (N^N tima)) (BEDI w+as) (VBN gecweden) (XP-PRD (PP (P under) <--- PP (NP (N .+a.)))) (PP (P o+d) (NP (NP-GEN (NPR^G Cristes)) (N tocyme) (PP (P on) (NP (N menniscnysse))))) (. ;)) (ID CathHomII,+ACHom_II,_12.1:110.8.2434)) (NODE (IP-SUB (NP-NOM *T*-1) (BEPI is) (VBN nemned) (XP-PRD (PP (P in) (NP (NPR Feppingum))))) (ID Bede3,Bede_3:15.222.31.826)) ( (CODE ) (IP-MAT (CONJ And) (PP (P for) (NP-INS (D^I +ti))) (NP-NOM (PRO^N heo)) (VBDI het) (IP-SMC (NP-ACC-SBJ (D^A +done) (N^A w+aterpytt)) (XP-PRD (PTP-NOM (PTP-NOM (VAG^G Libbendes)) (CONJP (CONJ &) (PTP-NOM (VAG^G Seondes) (NP (PRO Me))))))) (. ;)) (ID Heptateuch,Gen:16.14.628)) (NODE (IP-MAT (NP-NOM (NPR^N Hierusalem)) (BEPI is) (VBN gecweden) (NP-PRD (FW Visio) (FW pacis) <--- no XP with foreign words (, .) (IP-MAT-PRN (NP-NOM (D^N +t+at)) (BEPI is) (XP (NP-NOM (NP-GEN (N^G sibbe)) (N^N gesih+d)))))) (. ;)) (ID CathHomII,+ACHom_II,_4:37.235.874)) with appositives that are of a different category/case than their antecedent; note that foreign language appositives are labelled XP (in contradistinction to predicates (see above)) because they are not arguments (IP-MAT (ADVP-TMP (ADV^T +ta)) (VBD geecte) (NP-NOM (PRO^N he)) (ADVP-LOC (ADV^L +t+ar)) (PP (P to) (NP-ACC (D^A +t+at) (N^A word) (XP-PRN (NP-GEN (PRO$^G +tines) (N^G mu+des)))))) ( (CODE ) (IP-MAT (NP-NOM (D^N Se) (ADJ^N mildheorta) (NPR^N God)) (VBDI cw+a+d) (PP (P be) (NP-DAT (Q^D eallum) (ADJ^D synfullum) (N^D mannum))) (NP-ACC (NUM^A twa) (N^A word) (ADJP-ACC (ADV swi+de) (ADJ^A fremfulle)) (, ,) (XP-PRN (FW Declina) (FW a) (FW malo) (FW et) (FW fac) (FW bonum) (, ;) (IP-MAT-PRN (NP-NOM (D^N +t+at)) (BEPI is) (XP (IP-MAT (IP-MAT (VBI buh) (PP (P fram) (NP-DAT (N^D yfele)))) (CONJP (CONJ and) (IP-MAT (VBPH do) (NP-ACC (N^A god))))))))) (. .)) (ID AelfLives,+ALS[Ash_Wed]:145.1555)) in the +T+AT IS glossing construction to enclose the gloss, which may be of any category (IP, CP, PP, NP, etc.). When the gloss is nominal, the case of the gloss is either the same case as the element being glossed, or nominative; this gives rise to nom/acc ambiguity if the element being glossed is accusative. Therefore, case is only marked when it is unambiguous, or, if the element being glossed is nominative, the default is to label the gloss nominative as well. (NP-DAT (D^D +Disse) (N^D +teode) (, ,) (IP-MAT-PRN (NP-NOM (D^N +t+at)) (BEPI is) (XP (NP-DAT (NPR^D Nor+danhymbrum))))) <--- gloss with case matching (IP-SUB (NP-NOM *T*-1) (NP-ACC (D^A +tone) (ADJ^A o+terne) (N^A d+al) (NP-GEN (NP-GEN (NPR^G Nor+danhymbra)) (N^G riices)) (IP-MAT-PRN *ICH*-2)) (HVD h+afde) (, ,) (IP-MAT-PRN-2 (NP-NOM (D^N +t+at)) (BEPI is) (XP (NP (NPR Beornica))))) <--- ambiguous case (NP-ACC-TMP (NUM syx) (N^A ger) (ADJ^A ful) (, ,) (IP-MAT-PRN (NP-NOM (D^N +t+at)) (BEPI is) (XP (PP (P o+d) <--- non-nominal gloss (NP-DAT (N^D endan) (NP-GEN (NP-GEN (D^G +t+as) (N^G cyninges)) (N^G rices)))))))) ( (IP-MAT (CONJ and) (NP-NOM *con*) (HVD h+afde) (NP-DAT-ADT (PRO^D him)) (PP (P on) (NP-DAT (N^D mu+de))) (PP (P o+d) (NP-ACC (PRO$ his) (N^A for+dsi+d))) (NP-ACC (D^A +ta) (ADJ^A ylcan) (N^A word) (CP-REL (WNP-1 0) (C +te) (IP-SUB (NP *T*-1) (NP-NOM (D^N se) (ADJ^N eadiga) (NPR^N Stephanus)) (PP (P on) (NP (PRO$ his) (N ende))) (PP (P to) (NP-DAT (NPR^D Gode))) (VBDI gecw+a+d))) (, ;) (CODE ) (XP-PRN (FW Criste) (, .) (FW accipe) (FW spiritum) (FW meum) <--- parenthetical XP (, .) (IP-MAT-PRN (NP-NOM (D^N +t+at)) (BEPI is) (, .) (XP (IP-MAT (NP-NOM-VOC (NPR^N Crist)) <--- gloss XP (VBI onfoh) (NP-ACC (PRO$^A minne) (N^A gast))))))) (. .)) (ID CathHomII,+ACHom_II,_2:13.43.313))

Foreign phrases, non-argument (LATIN)

Foreign phrases which do not act as arguments are labelled LATIN and are not otherwise integrated in the clause. Note that speech is taken as a complement of verbs of saying, and thus foreign language speech is labelled as a quotation phrase (QTP) rather than LATIN. Latin appositives, as non-arguments, are labelled as X phrases.


(NODE (IP-MAT (PP (P On)
                  (NP-DAT (NP-GEN (D^G +tyses) (N^G cinges))
                          (N^D dagum)))
              (NP-NOM (NPR^N Laurentius) (N^N ercebiscop)
                      (CP-REL (WNP-NOM-1 (D^N se))
                              (C 0)
                              (IP-SUB (NP-NOM *T*-1)
                                      (BEDI was)
                                      (PP (P on)
                                          (NP-ACC (NPR^A Cent)))
                                      (PP (P +after)
                                          (NP-DAT (NPR^D Agustine))))))
              (RP+VBD for+tferde)
              (LATIN (FW iiii) (FW Nonae) (FW Februarii)))
      (ID ChronA,:616.8.286)) 

( (IP-MAT (CONJ &)
          (NP-ACC (PRO^A hine))
          (NP-NOM (MAN^N mon))
          (VBD gehalgode)
          (LATIN (FW IN) (FW UIGILIA) (FW Sancti) (FW Andree))
          (. .))
      (ID ChronA,:963.3.1395)) 

(NODE (IP-MAT (PP (P +Tonne)
                  (CP-ADV (C 0)
                          (IP-SUB (NP-NOM (D^N se) (N^N sacerd))
                                  (VBPI cristna+d))))
              (, ,)
              (ADVP-TMP (ADV^T +tonne))
              (VBPI or+da+t)
              (NP-NOM (PRO^N he))
              (PP (P on)
                  (NP-ACC (D^A +tone) (N^A man)))
              (, ,)
              (PP (P +tonne)
                  (CP-ADV (C 0)
                          (IP-SUB (NP-NOM (PRO^N hit))
                                  (ADVP (ADV swa))
                                  (VBPI gebyra+d))))
              (, ,)
              (LATIN (FW in) (FW modum) (FW crucis))
              (. ,))
      (ID WHom,WHom_8c:30.613)) 


(NP-ACC (D^A +da) (ADJ^A heofenlican) (N^A lofsangas)
	(XP-PRN (XP (FW Alleluian))
		(, .)
		(CONJP (CONJ and)
		       (XP (FW Gloria) (FW in) (FW excelsis) (FW deo)))))

( (IP-MAT (CONJ &)
          (NP-NOM (PRO$ his) (N^N lichama))
          (BEDI w+as)
          (VBN bebyriged)
          (PP (P on)
              (NP (NP-GEN (NPR Sancte) (NPR^G Petres))
                  (N cyricean)))
          (PP (P beforan)
              (NP-DAT (D^D +tam) (N^D husulportice)))
          (NP-INS-TMP (D^I +ty) (ADJ^I feor+tan) (N^I d+age)
                      (LATIN (FW idus) (FW Martiarum)))
          (. ,))
      (ID Bede2,Bede_2:1.94.6.7))

Isolated latin verses (such as those introducing homilies, etc.) are treated as separate tokens and labelled LATIN. LATIN generally only dominates foreign words (POS tag FW), but glosses of the +T+AT IS type, if present, are included in the parens labelled LATIN (no examples yet).


( (CODE )
  (LATIN (FW XIII.) (FW KALENDAS) (FW FEBRUARII)      <--- isolated verse
         (. .)) (ID AelfLives,+ALS_[Sebastian]:1.2))
( (CODE )
  (LATIN (FW PASSIO) (FW SANCTI) (FW SEBASTIANI) (FW MARTYRIS)
         (. .)) (ID AelfLives,+ALS_[Sebastian]:1.3))

Time expressions with +T+AS

Time expressions involving +T+AS are parsed as follows:

+T+AS alone or modified by an adverb (usually SONA or RA+DE) (NP-GEN-TMP (ADVP (ADV sona/ra+de)) (D^G +t+as)) +T+AS + PP/NP (PP (NP-GEN (D^G +t+as)) (P ymb) (NP-ACC iii niht)) (NP-INS-TMP (NP-GEN (D^G +t+as)) (D^I +ty) (ADJ^I +afterran) (N^I geare)) PP/NP + +T+AS (PP (P ymb) (NP (NUM^A +treo) (N^A niht) (NP-GEN (D^G +t+as)))) (NP-INS-TMP (D^I +Ty) (ADJ^I +afterran) (N^I geare) (NP-GEN (D^G +t+as))) PP/NP + +T+AS +TE + clause The +T+AS +TE clause is treated as a time relative. (PP (P ymb) (NP-ACC (NUM x) (N^A gear) (CP-REL (WNP-GEN-1 (D^G +t+as)) (C +te) (IP-SUB (NP-GEN-TMP *T*-1) he us fulwiht send)))) (NP-TMP (NUMP (NUM ccc) (CONJ &) (NUM xcvi)) (NP-GEN (N^G wintra)) (CP-REL (WNP-GEN-1 (D^G +t+as)) (C +te) (IP-SUB (NP-GEN-TMP *T*-1) ...)))

NPs of extent (-EXT)

NPs of extent occur only as modifiers of other categories, never at IP-level; NPs of extent at IP-level are labelled as ordinary adjuncts. NPs of extent can modify quantifiers, adjectives, adverbs and PPs. The most common types are:

instrumental determiners (+TY (usually spelled +TE) and +TON) and quantifiers (MYCLE, etc) in the meaning something like to that extent/to a great extent modifying quantifiers (usually MA/L+AS), adjectives and adverbs ( (IP-MAT (CONJ &) (NP-NOM *exp*) (QP-ADT (NP-INS-EXT (Q^I mycle)) <--- Q modifying Q (Q l+as)) (NP-DAT (D^D +tam) (CP-REL *ICH*-1)) (BEDI w+as) (VBN alyfed) (, ,) (CP-REL-1 (WNP-DAT-2 (D^D +tam)) (C +te) (IP-SUB (NP-NOM *T*-2) (ADVP (ADV niwan)) (VBDI com))) (. ,)) (ID GDC1,:4.28.34.227)) (NODE (IP-SUB (NP-NOM (PRO^N heo)) (NEG ne) (BEPI by+t) (ADVP (NP-INS-EXT (D^I +ty)) <--- D modifying ADV (ADV hra+tor)) (RP onweg) (VBN adrifen) (PP (P fram) (NP (NP-GEN (D^G +t+are) (N^G heortan)) (N +amtignesse)))) (ID GDC1,:4.35.12.304)) (NODE (IP-MAT (ADVP (ADV Witodlice)) (NP-GEN-TMP (D^G +t+as)) (QP-ADT (NP-INS-EXT (D^I +te)) <--- D modifying Q (Q ma)) (NP-NOM (D^N seo) (N^N spr+ac)) (PP (P be) (NP-DAT (PRO^D him))) (VBD ferde) (. FULL-STOP)) (ID GospLk,:5.15.368)) Not uncommonly, +T+AS is present as well in this construction. While it is not clear what the internal structure of these phrases is, the +T+AS seems to modify the rest of the phrase and thus is represented as a sister. ( (IP-MAT-SPE (CONJ and) (NP-NOM (PRO^N ic)) (BEPS beo) (ADJP-NOM-PRD (NP-GEN (D^G +t+as)) (ADJP-NOM (NP-INS-EXT (D^I +te)) <--- +T+AS +TE (ADJ^N geleaffulra))) (NP-DAT (D^D +tam) (VAG^D lifigendan) (NPR^D Gode)) (NP-DAT-TMP (Q^D eallum) (PRO$^D minum) (N^D dagum)) (PP (P gif) (CP-ADV-SPE (C 0) (IP-SUB-SPE ...)))) (ID AelfLives,+ALS_[Swithun]:331.2567)) measures of time or distance modifying adverbs and PPs; note that especially with PPs the NP of extent may follow the PP rather than precede it ( (IP-MAT (CONJ &) (ADVP-TMP (NP-INS-EXT (NUM^I ane) (N^I geare)) <--- modifying ADV (ADV^T +ar)) (NP-NOM (PRO$ his) (N^N bro+dur) (CP-REL *ICH*-1)) (RP+VBD for+tferde) (, ,) (CP-REL-1 ...)) (ID ChronA,:885.13.943)) (NODE (IP-SUB (NP-NOM *T*-2) (NP-ACC (NPR^A Iacobum)) (RP+VBDI ofslog) (PP (NP-INS-EXT (NUM^I ane) (N^I geare)) <--- modifying PP (P +ar) (NP-DAT (PRO$ his) (ADJ^D agnum) (N^D dea+te)))) (ID ChronA,:46.1.74)) (NODE (IP-MAT (PP (P AER) (NP (NP-GEN (NPR^G Cristes)) (N gefl+ascnesse)) (NP-EXT (NUM lx) <--- modifying PP, following (NP-GEN (N^G wintra)))) (, .) (NP-NOM (NPR^N Gaius) (NPR^N Iulius) (NP-NOM-PRN (D^N se) (N^N Casere))) (ADVP-TMP (ADV^T +arest) (NP-GEN (NPR^G Romana))) (NP-ACC (NPR^A Bretenlond)) (VBD gesohte) (. .)) (ID ChronA,:0.45.51)) (NODE (IP-MAT (PP (P +Ar) (D^D +dam) (CP-ADV (C +de) (IP-SUB (NP-NOM (NPR^N Romeburh)) (VBN getimbred) (BEDS w+are))) (NP-EXT (NUMP (NUM eahta) (NUM $hund)) (NP-GEN (N^G wintra)))) (, ,) (PP (P mid) (NP-DAT (NPR^D Egyptum))) (BEDI wear+d) (NP-ACC-TMP (NUM syfan) (N^A gear)) (NP-NOM (D^N se) (ADJ^N ungemetlica) (N^N eor+dwela)) (. ,)) (ID Orosius,:5.23.19.362)) (NODE (IP-MAT (PP (P On) (NP-INS (D^I +ty) (ADJ^I ylcan) (N^I gere))) (VBD worhte) (NP-NOM (D^N se) (RP+VBN^N foresprecena) (N^N here)) (NP-ACC (N^A geweorc)) (PP (P be) (NP (NPR Lygan))) (PP (NP-EXT (NUM xx) <--- distance modifying PP (NP-GEN (N^G mila))) (P bufan) (NP-DAT (NPR^D Lundenbyrig))) (. .)) (ID ChronA,:896.1.1098)) (NODE (IP-SUB (NP-NOM *T*-3) (BEDI w+as) (PP (NP-DAT-EXT (NUM VIII) (N^D milum)) (P fram) (NP (PRO$ his) (N byrig)))) (ID GDC3a,:11.194.17.306)) measures of distance modifying adjectives like BRAD, LANG, WIDE, etc. These are generally genitive when it is possible to tell. ( (IP-MAT (NP-NOM (D^N se) (N^N wudu)) (BEPI is) (ADVP-DIR (ADV^D eastlang) (CONJ &) (ADV^D westlang)) (ADJP-NOM-PRD (ADJP-NOM (NP-GEN-EXT (NUM^G hundtwelftiges) (NP-GEN (N^G mila))) (ADJ^N lang) (CONJP (CONJ o+t+te) (ADJX (ADJ^N lengra)))) (, ,) (CONJP (CONJ &) (ADJP-NOM (NP-GEN-EXT (NUM^G +tritiges) (NP-GEN (N^G mila))) (ADJ^N brad)))) (. ;)) (ID ChronA,:893.5.1011)) (NODE (IP-MAT (CONJ &) (NP-NOM (D^N +d+at) (N^N w+ater)) (BEDI w+as) (ADJP-NOM-PRD (NP-GEN-EXT (NUM fyftyne) (N^G f+a+dma)) (ADJ^N deop)) (PP (P ofer) (NP-ACC (D^A +da) (ADJ^A heahstan) (N^A duna))) (. .)) (ID Heptateuch,:7.20.318)) A single +T+AS (without +TE/+TY) modifying an adjective is taken as an extent item. (ADJP (NP-GEN-EXT (D^G +t+as)) (ADJ deop)) In EALLES SWA X, EALLES is marked as an extent item ( (IP-MAT (CONJ &) (NP-NOM (N^N men)) (ADVP-TMP (ADV^T +tonne)) (VBPI lufia+d) (, ,) (IP-MAT-PRN (NP-NOM (PRO^N he)) (VBDI cw+a+d)) (, ,) (ADVP (QP-GEN-EXT (Q^G ealles)) (ADV to) (ADV swy+de)) (NP-ACC (D^A +tas) (ADJ^A swicolan) (N^A woruld)) (. FULL-STOP)) (ID WHom,WHom_5:16.171))

Interjection Phrases (INTJP)

Single words tagged INTJ are not given an INTJP label. The INTJP label is used in three cases.

When a single word not POS-tagged INTJ, such as HW+AT, is used as an interjection ( (CODE ) (IP-MAT (INTJP (WPRO Hw+at)) (, ,) (ADVP-TMP (ADV^T +da)) (NP-NOM (NPR^N Sem) (CONJ &) (NPR^N Iapheth)) (VBDI dydon) (NP-ACC (NUM^A anne) (N^A hwitel)) (PP (P on) (NP-DAT (PRO$ heora) (N^D sculdrum))) (. ,)) (ID Heptateuch,Gen:9.23.409)) To group two or more words used as an interjection; one or more of these may, but doesn't have to be tagged INTJ ( (IP-MAT (CONJ &) (NP-NOM *con*) (ADVP (ADV eorringa)) (ADVP (ADV +tus)) (VBDI cw+a+d) (. :)) (ID GDC1,:9.62.16.625)) ( (INTJP (INTJ wala) (INTJ wa) ( (IP-MAT (CONJ &) (NP-NOM *con*) (VBDI cw+a+d) (PP (NP-DAT (PRO^D him)) (P to)) (, :) (CP-QUE-SPE (INTJP (WPRO Hw+at) (INTJ la)) (, ,) (WADVP-1 (WADV hwi)) (IP-SUB-SPE (ADVP *T*-1) (VBDI dydest) (NP-NOM (PRO^N +du)) (ADVP (ADV swa)) (PP (P wi+d) (NP (PRO me))))) (. ;)) (ID Heptateuch,Gen:12.18.494)) ( (IP-MAT-SPE (INTJP (ADV Ono) (WPRO hw+at)) (NP-NOM (PRO^N +tu)) (ADVP-TMP (ADV^T nu)) (HVPI hafast) (PP (P +turh) (NP (NP-GEN (NPR^G Godes)) (N gife))) (NP-ACC (NP-GEN (PRO$^G +tinra) (N^G feonda) (CP-REL-SPE *ICH*-1)) (N^A hond)) (VBN^A (VBD beswicade)) (, ,) (CP-REL-SPE-1 (WNP-ACC-2 (D^A +ta)) (C 0) (IP-SUB-SPE (NP-ACC *T*-2) (NP-NOM (PRO^N +du)) (NP-RFL-ADT (PRO +de)) (RP+VBD ondrede))) (. ,)) (ID Bede2c,:9.132.22.60))

WH-phrases

See the POS manual for some differences in the labelling of wh-words from the PPCME2.

Note especially that while (the very few) wh-quantifiers in the PPCME2 were labelled as wh-words, in the YCOE they are labelled as quantifiers.

See also WHETHER and Wh-operators and traces.

Conjunction

Conjunction of unlike constituents
EAC, SWYLCE, SAMOD, etc.
+AG+TER GE...GE, O+TER +TE...+TE
HW+A+TER...+TE...+TE
Conjoined structures with NEALLES and/or NA
NEALLES +T+AT AN +T+AT...
Elision
Elision in conjoined clauses Elision in subordinate clauses under identity with the matrix Elision in NPs and other phrases
PPCME2 Table of contents
YCOE Table of contents

The YCOE uses the same scheme for conjunction as the PPCME2. Note that the way conjunction is annotated has especially important consequences for locating shared (or potentially shared) modifiers.

Conjunction of unlike constituents

Although in general when two unlike constituents are conjoined the dominating phrase will simply be the first of the two, there are some special cases which arise because of the way in which POS tags are assigned.


(PP (PP ...)
    (CONJP (CONJ ond)
	   (ADVP ...)))

(ADVP (ADVP ...)
      (CONJP (CONJ ond)
	     (PP ...)))

The special cases involve subcategories of adjectives and verbs. Some adjectives (MICEL and LYTEL) are for consistency reasons tagged as quantifiers; in conjunction they are treated as (although they are not tagged as) adjectives and conjoin freely with other adjectives under an ADJP label.


(ADJP-DAT (Q^D myclum) (CONJ &) (ADJ^D hefegum))

(NP-NOM (ADJP-NOM (Q^N mycel)
		  (CONJP *ICH*-1))
	(N^N wol)
	(CONJP-1 (CONJ &) (ADJ^N grim)))

(NP-DAT (ADJP-DAT (QP-DAT (ADV swa) (Q^D miclum))
		  (CONJP *ICH*-1))
	(N^D were)
	(CONJP-1 (CONJ &)
		 (ADJP-DAT (ADV swa) (ADJ^D halgum))))))

Likewise participles are often adjectival and in conjunction with adjectives are treated as adjectives. This includes participle phrases.


(ADJP-NOM (ADJ^N snotor) (CONJ and) (VBN^N gelyfed))

(ADJP-NOM-PRD (ADJ^N eadmod) (CONJ and) (VBN ge+tungen))

(ADJP-NOM (PTP-NOM (PP (P on)
		       (NP (N godnysse)))
		   (VAG^N scinende))
	  (CONJP (CONJ and)
		 (ADJP-NOM (PP (P on)
			       (NP-DAT (Q^D eallum) (N^D +teawum)))
			   (ADJ^N arwur+dful))))

Because verbs are tagged according to form not function, it is often the case that a formally ambiguous verb (VBD, VBP, etc.) is conjoined to a verb specified for mood (VBDI, VBDS, VBPS, VBPI, etc.). In this case, the more specific tag takes precedence and dominates both forms; but note that cliticized particles and negation are never carried up to the dominating node. (See Word-level conjunction).


(VBDI (VBD andwyrde) (CONJ and) (VBDI cw+a+d))

(VBDS (VBDS woosce) (CONJ &) (VBD cl+ansode))

(VB (RP+VB onfoon) (CONJ &) (VB +ticgan))

EAC, SWYLCE, SAMOD, etc.

When necessary, EAC and SWYLCE are treated as part of a CONJP, although they are still labelled ADV. This is generally the case when they are used in conjoining phrases smaller than IP, or when conjoining CPs. When IPs are conjoined, the default is to make EAC and SWYLCE part of the IP rather than the CONJP.


(NP-ACC (NP-ACC (D^A +ta) (ADJ^A sylfrenan))
	(CONJP (CONJ and) (ADV eac) (ADV swilce)
	       (NP-ACC (D^A +da) (ADJ^A gyldenan))))

(NP-DAT (NP-DAT (PRO$^D eowrum) (N^D cynedome))
	(CONJP (CONJ and) (ADV eac)
	       (NP-DAT (PRO$^D eowrum) (N^D folce)))))

(IP-SUB (IP-SUB (NP-NOM (PRO^N heo))
		(BEDS w+are)
		(PP (P to)
		    (NP (N bisene)))
		(VBN asteald)
		(NP-DAT-ADT (Q^D eallum) (N^D middangearde)))
	(, ,)
	(CONJP (CONJ &)
	       (IP-CON (ADVP (ADV eac))
		       (ADVP (ADV swelce))
		       (NP-NOM (PRO^N heo) 
			       (ADJP-NOM (ADJ^N self)))
		       (VAG sprecende)
		       (BEPS sie)
		       (PP (P to)
			   (NP-DAT (Q^D eallum) (N^D moncynne))))))

A following EAC, or other adverb meaning more or less together, (as for example, SAMOD, TOSAMNE, ENDEMES, +ATG+ADERE, TOG+ADERE, TOEACAN) or MID, which in this context is tagged as an adverb rather than a particle is treated in the same way; that is, it is put at CONJP-level, and does not project a phrase.


(NP-NOM (NP-NOM (D^N +t+at) (N^N feoh))
	(CONJP (CONJ and)
	       (NP-NOM (NUM twentig)
		       (NP-GEN (N^G penega)))
	       (ADV toeacan)))

(NP-TMP (NP (NUM feowertig)
	    (NP-GEN (N^G daga)))
	(CONJP (CONJ &)
	       (NP (NUM feowertig)
		   (NP-GEN (N^G nihta)))
	       (ADV tog+adere)))

(NP-ACC (NP-ACC (D^A +da) (N^A wylne)
		(NP-ACC-PRN (NPR^A Agar)))
	(CONJP (CONJ &)
	       (NP-ACC (NPR^A Ismahel))
	       (ADV samod))))

(NP-NOM-PRN-1 (NP-NOM (N^N eorlas) (ADJ^N anmode))
	      (, ,)
	      (CONJP (CONJ ond)
		     (NP-NOM (PRO$ hira) (N^N idesa))
		     (ADV mid)))

+AG+TER GE...GE, O+TER +TE...+TE

Sequences of +AG+TER GE...GE and O+TER +TE...+TE are treated as multi-word conjunctions, whether the first two parts appear together or separately, except when the first element has overt inflection. In the latter case, the +AG+TER/O+TER part is treated as a complement of the verb and the conjuncts as appositives on it.


(NP-NOM-PRN (CONJ +ag+der) (CONJ ge)
	    (NP-NOM (N^N inbyrdlingas)) 
	    (CONJP (CONJ ge)
		   (NP-NOM (VBN^N gebohte) (N^N +deowan)))
 	    (CONJP (CONJ &)
		   (NP-NOM (ADJ^N +al+deodige) (N^N men))))

(NP-ACC (CONJ +ag+der) (CONJ ge) (N^A god) (CONJ ge) (N^A yfel))

(PP (CONJ +ag+der) (CONJ ge)
    (PP (P on)
	(NP-DAT (NPR^D Sunnondagum)))
    (CONJP (CONJ ge)
	   (PP (P on)
	       (NP-DAT (ADJ^D o+drum) (N^D dagum)))))

*difference*
Note that in YCOE conjunctions are not traced back to their logical position when they have floated leftward as they are in the PPCME2.

HW+A+TER...+TE...+TE

When alternatives are given in a WHETHER question, they are generally conjoined by +TE...+TE, but sometimes by +TE...HW+A+TER +TE. When the conjoined constituents are clausal, an embedded HW+A+TER of this sort is labelled WQ and the whole treated as a conjoined question; when the constituents are less than clausal, however, the embedded HW+A+TER is simply labelled as a conjunction. See also WHETHER questions.


( (CODE )
  (CP-QUE-SPE (WQ Hw+a+der)
              (IP-SUB-SPE (BEDI w+as)
			  (NP-NOM (NP-GEN (NPR^G Iohannes))
				  (N^N fulluht))
			  (PP (CONJ +te)
			      (PP (P of)
				  (NP-DAT (N^D heofonum)))
			      (CONJP (CONJ +te)
				     (PP (P of)
					 (NP-DAT (N^D mannum))))))
              (. ;)) (ID WSGospels,Mt_[WSCp]:21.25.1416))

( (CODE )
  (CP-QUE-SPE (INTJ Eala)
              (NP-NOM (PRO^N ge)
                      (ADJP-NOM (ADJ^N blindan)))
              (, ,)
              (WQ hw+a+ter)
              (IP-SUB-SPE (BEPI ys)
			  (QP-NOM-PRD (Q^N mare))
			  (NP-NOM (CONJ +te) 
				  (NP-NOM (N^N offrung))
				  (CONJP (CONJ +te)
					 (NP-NOM (D^N +t+at) (N^N weofud)
						 (CP-REL-SPE (WNP-NOM-1 0)
							     (C +te)
							     (IP-SUB-SPE (NP-NOM *T*-1)
									 (VBPI gehalga+d)
									 (NP-ACC (D^A +ta) (N^A offrunge))))))))
              (. ?)) (ID WSGospels,Mt_[WSCp]:23.19.1582))

( (CP-QUE-SPE (WPP-1 (P +at) 
                     (WNP-DAT (WPRO^D hwam)))
              (IP-SUB-SPE (PP *T*-1)
			  (VBPI nima+d)
			  (NP-NOM (N^N cyningas))
			  (NP-ACC (N^A gafol) (CONJ o+d+de) (N^A toll))
			  (, ,)
			  (PP-PRN (PP (P of)
				      (NP-DAT (PRO$ hyra) (N^D bearnum)))
				  (CONJP (CONJ hw+a+der) (CONJ +de) <--- HW+A+TER labelled as a conjunction
					 (PP (P of)
					     (NP-DAT (N^D fremedum))))))
              (. ?)) (ID WSGospels,Mt_[WSCp]:17.25.1152))

Conjoined structures with NEALLES and/or NA

X NEALLES Y, X NEALLES NA Y and X NA Y have an empty CONJP and NEALLES/NA are part of the second conjunct. Cf. X NOT Y conjunction in the PPCME2.


(NP (NP (N firstconjunct))
    (CONJP (NP (NEG+Q+G nealles) (NEG+ADV na)
               (N secondconjunct))))

NEALLES (NA) X AC Y is done like NOT X BUT Y in the PPCME2.


(NP (NP (NEG+Q+G nealles) (NEG+ADV na) 
	(N firstconjunct))
    (CONJP (CONJ ac)
           (NP (N secondconjunct))))

NEALLES +T+AT AN +T+AT...

The words NEALLES +T+AT AN +T+AT for Latin non solum ... (et etiam) defy grammatical description. The parts of this string are therefore given literal PoS tags and simply enclosed in an ADVP. The second +T+AT is labelled C because it seems to cooccur with clauses although it doesn't always act as a subordinator. If the two clauses joined are main clauses, the NALES clause is separated from the AC clause, since both contain a matrix verb.


(ADVP (NEG+Q^G nealles) (D +t+at) (NUM an) (C +t+at))

Elision

Elision in conjoined clauses
Elision in subordinate clauses under identity with the matrix
Elision in NPs and other phrases

PPCME2

Elision in conjoined clauses

We go to some trouble in the YCOE to try to ensure that it is possible to differentiate between clauses which are essentially complete and those in which core material (verbs and subjects) has been elided. When only the subject is elided, an empty subject is added. When verbal material is missing due to the elision of various elements under identity with another clause within the same token the clause has an index beginning with =0; the complete clause on which it is patterned has a matching -0 index.


( (IP-MAT (IP-MAT-0 (CONJ and)
                    (NP-NOM *con*)
                    (PP (P mid)
                        (NP-DAT (Q^D mycclum) (N^D heofungum)))
                    (NP (PRO$ heora) (N geleafan))
                    (MDDI woldon)
                    (VB awendan))
          (CONJP (CONJ and)
                 (IP-MAT=0 (NP-ACC (D^A +ta) (N^A wita))
			   (VB gestillan)))
          (. .)) (ID AelfLives,+ALS_[Sebastian]:34.24))

Between two clauses of the same type (both matrix or both subordinate) this system is used for elision under IP and VP conjunction, which are not differentiated because of the lack of a VP node in the annotation system. Both clauses are labelled IP-MAT (or IP-SUB) plus the appropriate index. They are conjoined, if an overt conjunction is present, or simply parallel.


( (CODE )
  (IP-MAT (IP-MAT-0 (NP-NOM (PRO^N Hi))  <--- conjoined clauses with elision
                    (NEG ne)
                    (MDPI magon)
                    (NP-DAT-RFL (PRO^D him) 
                                (ADJP-DAT (ADJ^D sylfum)))
                    (VB fremian))
          (, ,)
          (CONJP (NEG+CONJ ne)       
                 (IP-MAT=0 (NP-DAT (ADJ^D o+drum))
                            (VB fultumian)))
          (. .)) (ID AelfLives,+ALS[Agnes]:206.643))

(CP-THT (C +t+at)
	(IP-SUB (IP-SUB-0 (NP-NOM (PRO^N ic))
			  (NP-DAT (PRO$^D +dinum) (ADJ^D awyrgedum) (N^D godum))
			  (NEG ne)
			  (MDPS +durfe)
			  (VB geoffrian))
		(, ,)
		(CONJP (NEG+CONJ ne)
		       (IP-CON=0 (PP (P +turh)
				     (NP-ACC (N^A +alfremede) (N^A horwan)))
				 (ADVP-TMP (ADV^T +afre))
				 (BE beon)
				 (VBN gefyled)
				 (PP (P mid)
				     (NP-DAT (D^D +tam) (ADJ^D fulum)
					     (N^D myltestrum)))))))


( (IP-MAT (IP-MAT-0 (NP-NOM (ADJ^N o+der))  <--- parallel clauses with elision
                    (BEDI w+as)
                    (ADJP-NOM-PRD (ADJ^N w+aterseoc)))
          (, ,)
          (IP-MAT=0 (NP-NOM (ADJ^N o+der))  
                    (QP (Q eall))
                    (PP (P on)
                        (NP-DAT (N^D wundum))))
          (. ;))
      (ID AelfLives,+ALS_[Sebastian]:144.87))

A similar system is used for cases of right-node raising and various internal parenthetical constructions involving elision. Here the second conjunct is also given a -PRN label (e.g., IP-MAT-PRN=0) and put at IP-level. This essentially removes it from the clause allowing the following material to be interpreted with the first conjunct.


( (CODE )
  (IP-MAT-0 (NP-NOM (D^N Se) (N^N cniht))   <--- right-node raising
            (BEDI wear+d)
            (VBN geancsumod)
            (IP-MAT-PRN=0 (CONJ and) 
			  (ADVP-LOC (ADV^L wi+dinnan))
			  (VBN ablend))
            (PP (P +after)
                (NP (NP-GEN (D^G +t+as) (N^G m+adenes)
                            (CP-REL *ICH*-1))
                    (N spr+ace)
                    (, ,)
                    (CP-REL-1 (WNP-NOM-2 0) 
                              (C +te)
                              (IP-SUB (NP-NOM *T*-2)
                                      (NP-ACC (PRO^A hine))
                                      (VBDI spearn)
                                      (PP (P mid)
                                          (NP-DAT (N^D wordum)))))))
            (. .)) (ID AelfLives,+ALS[Agnes]:63.547))

( (IP-MAT-0 (CONJ ac)
            (NP-NOM (NUM^N twegen) 
                    (NP-GEN (PRO$ his) (N^G geferan))) 
            (VBDI feollon)
            (PP (P be)
                (NP-DAT (N^D wege)))
            (, ,)
            (IP-MAT-PRN=0 (NP-NOM (NUM^N an))   <--- parenthetical clause with elision
			  (PP (P of)
			      (NP-DAT (NUM^D anum) (N^D stypele))))
            (, ,)
            (IP-MAT-PRN=0 (NP-NOM (ADJ^N o+ter))
			  (PP (P on)
			      (NP-ACC (NUM^A anne) (N^A stan))))
            (, ,)
            (PP (P swa)
                (CP-ADV (C +t+at)
                        (IP-SUB (IP-SUB (NP-NOM (PRO^N hi))
                                        (BEDI wurdon)
                                        (RP+VBN^N tocwysede))
                                (CONJP (CONJ and)
                                       (IP-CON (NP-NOM *con*) 
                                               (PTP-NOM (VAG^N cwylmiende))
                                               (VBDI lagon))))))
            (. ,)) (ID AelfLives,+ALS_[Maur]:90.334))

Elision in subordinate clauses under identity with the matrix

In addition to indicating elision under conjunction in parallel clauses, the equal-sign coindexing system is used to link matrix and subordinate clauses when material has been elided from the subordinate clause under identity with the matrix. (*difference* Note that in the PPCME2 this kind of elision is indicated by an empty verb (VB *), which can also stand for the whole VP.)


( (IP-MAT-0 (CONJ and)
            (NP-NOM *con*)
            (VBD eode)
            (PP (P to)
                (NP (D +t+are) (N ea)))
            (PP (P +ta+da)
                (CP-ADV (C 0)
                        (IP-SUB=0 (NP-NOM (PRO^N he))
                                   (PP (P ofer)
                                       (NP-ACC (D^A +ta) (N^A brycge)))
                                   (NEG ne)
                                   (MDD mihte))))
            (. ,)) (ID AelfLives,+ALS_[Alban]:93.2240))

In addition there are specific rules for elision in comparative clauses, which suffer from more extensive and problematic elision than other types of clauses.

Elision in NPs and other phrases

In conjoined NPs the head noun may be elided in the second conjunct. We do not explicity represent this in the annotation.


(NP-ACC (NP-ACC (D^A +tone) (ADJ^A so+dan) (NPR^A God))
	(CONJP (CONJ &)
	       (NP-ACC (D^A +tone) (VAG^A lifigendan)))) <--- GOD elided

(NP-ACC (NP-ACC (D^A +ta) (ADJ^A elreordan) (N^A +teode))
	(CONJP (CONJ &)
	       (NP-ACC (D^A +ta) (ADJ^A re+dan)))        <--- +TEODE elided
	(CONJP (CONJ &)
	       (NP-ACC (D^A +ta) (ADJ^A ungeleafsuman))))

Likewise, the head may be missing in the first conjunct, due to right-node raising. Again this is not explicitly represented.

(NP-ACC (NP-ACC (D^A +ta) (ADJ^A orsorgan))   <--- right-node raising of BLYSSE
	(CONJP (CONJ and)
	       (NP-ACC (D^A +ta) (ADJ^A unateorigendlican) (N^A blysse))))

(NP-NOM (NP-NOM (D^N +da) (ADJ^N Romaniscan))
	(CONJP (CONJ and)
	       (NP-NOM (D^N +da) (ADJ^N re+dan) (ADJ^N Iudeiscan) (Q^N manega))))

Similarly, in conjoined PPs, the head of the NP complement of the second PP conjunct is sometimes elided. When the remaining element is an modifier, this is treated in the usual way, by simply ignoring the missing head. When it is a complement, as for example a genitive, an extra NP node is included to indicate that the preposition doesn't take the genitive directly.


(PP (CONJ o+d+to)
    (PP (P in)
	(NP (NP-GEN (NPR^G Romana))
	    (N cirican)))
    (CONJP (CONJ o+d+to)
	   (PP (P in)
	       (NP (NP-GEN (NPR^G Gallia)))))    <--- CIRICAN elided
    (CONJP (CONJ o+d+to)
	   (PP (P in)
	       (NP-DAT (Q^D hwylcre) (ADJ^D o+derre)))))

(PP (PP (P to)
	(NP (NP-GEN (NPR^G Cantwara))
	    (N rice)))
    (, ,)
    (CONJP (CONJ &)
	   (PP (P to)
	       (NP (NP-GEN (NPR East) (NPR^G Seaxna))
		   (N rice))))
    (, ,)
    (CONJP (CONJ &)
	   (PP (P to)
	       (NP (NP-GEN (NPR^G Su+trigea)))))  <--- RICE elided
    (, ,)
    (CONJP (CONJ &)
	   (PP (P to)
	       (NP (NP-GEN (NPR Su+t) (NPR^G Seaxna))
		   (N rice))))))

Empty categories

Empty subjects
WH-operators and traces
Position of traces
Empty quantifier heads
PPCME2 Table of contents
YCOE Table of contents

PPCME2

Empty subjects

The following three types of empty subjects are used in YCOE, all defined in the same way as in the PPCME2.

subjects elided under conjunction (NP-NOM *con*) small pro subjects (NP-NOM *pro*) expletive subjects (NP-NOM-x *exp*)

Arbitrary PRO (*arb*) is not used; arbitrary subjects in the accusative and infinitive construction are not indicated.

WH-operators and traces

PPCME2

The only difference between the PPCME2 and YCOE regarding traces is the addition of case in the YCOE. The following rules apply for marking case on empty categories:

non-overt wh-phrases and their traces are only labelled for case when they are nominative (as either subject or predicate); for predicates, this applies mostly to wh-adjectives in comparatives ( (IP-MAT-SPE (CONJ and) (NP-NOM *con*) (NEG ne) (MDPI cunnon) (NP-ACC (D^A +t+at) (ADJ^A o+der) (CP-REL-SPE (WNP-NOM-1 0) <--- case on subject wh- (C +te) (IP-SUB-SPE (NP-NOM *T*-1) (ADVP-TMP (ADV^T +afre)) (ADJP-NOM-PRD (ADJ^N endeleas)) (BEPI bi+d)))) (. .)) (ID AelfLives,+ALS_[Sebastian]:61.40)) ( (IP-MAT (CONJ and) (NP-NOM *con*) (NP-ACC (D^A +ta) (CP-REL *ICH*-1)) (NP-DAT (NPR^D Gode)) (VBD gebrohte) (CP-REL-1 (WNP-2 0) <--- no case on non-subject wh- (C +te) (IP-SUB (NP *T*-2) (NP-NOM (D^N se) (N^N deofol)) (VB +atbredan) (MDD wolde))) (. .)) (ID AelfLives,+ALS_[Sebastian]:21.17)) (ADJP-NOM-PRD (ADV swa) (ADJ^N hal) (PP (P swa) (CP-CMP-SPE (WADJP-NOM-1 0) <--- non-overt WADJP-NOM in comparative (C 0) (IP-SUB-SPE=0 (ADJP-NOM-PRD *T*-1) (NP-NOM (PRO^N ic)))))) (ADJP-NOM-PRD (ADJ^N deorwur+dra) (PP (P +tonne) (CP-CMP (WADJP-NOM-1 0) (C 0) (IP-SUB=0 (ADJP-NOM-PRD *T*-1) (NP-NOM (D^N +da) (ADJ^N dyran) (N^N ma+dmas)))))) (NP-NOM (ADJ^N swilce) (N^N halgan) (PP (P swylce) (CP-CMP (WNP-NOM-1 0) <--- non-overt WNP-NOM in comparative (C 0) (IP-SUB (NP-NOM-PRD *T*-1) (NP-NOM (D^N +t+as) (ADJ^N halga) (N^N cyning)) (BEPI is))))) wh-phrases, overt or empty, are given function labels when the function labels are lexically based; that is, -TMP, -LOC, -DIX, -DIR, ( (IP-MAT (CONJ and) (NP-NOM *con*) (VBD s+ade) (CP-QUE (WADVP-LOC-1 (WADV^L hw+ar)) (C 0) (IP-SUB (ADVP-LOC *T*-1) (NP-NOM (PRO$ his) (N^N lichama)) (VBDI l+ag) (PP (P on) (NP-DAT (D^D +tam) (N^D adelan))))) (. ,)) (ID AelfLives,+ALS_[Sebastian]:461.281)) ( (CODE ) (IP-MAT (NP-NOM (D^N Seo) (ADJ^N eadiga) (NPR^N Lucia)) (PP (P on) (NP-DAT (D^D +t+are) (ADJ^D ylcan) (N^D stowe) (CP-REL *ICH*-1))) (VBD wunode) (CP-REL-1 (WADVP-LOC-2 0) (C +te) (IP-SUB (ADVP-LOC *T*-2) (NP-NOM (PRO^N heo)) (RP+VBN ofslagen) (BEDI w+as))))) (QP-ADT (QP-INS-EXT (ADV swa) (Q^I mycele)) (Q ma) (CP-CMP (WQP-1 0) (C 0) (IP-SUB (QP-ADT *T*-1) (NP-NOM (PRO^N he)) (VBPI gefealh) (PP (P mid) (NP (N geornnysse))) (NP-DAT (D^D +tam) (N^D gebedum))))) overt wh-phrases may be given both case and function labels when appropriate (PP (P ymbe) (NP (NUM xli) (NP-GEN (N^G wintra)) (PP (P butan) (NP-DAT (NUM^D anre) (N^D niht))) (NP-GEN (D^G +t+as) (CP-REL (WNP-GEN-1 +t+as) (C +te) (IP-SUB (NP-GEN-TMP *T*-1) (NP-NOM (NPR^N +Alfred) (N^N cyning)) (RP+VBD for+tferde)))))) The trace of a non-overt wh-phrase is given the appropriate function label for its role in the clause if this is clear, but not case (IP-SUB-SPE (NP-NOM (D^N seo) (ADJ^N uplice) (N^N arf+astnis)) (NP (PRO me)) (NP-ACC (Q^A +anig) (N^A f+ac) (CP-EOP-SPE (WNP-1 0) (IP-INF-SPE (NP-ADT *T*-1) (TO to) (VB^D lifigenne)))) (VB forgeofan) (MDP wille))

Position of traces

Traces always appear at the edge of phrases, at the left edge if the antecedent precedes the trace (as with all wh-traces and some A'-movement traces, as for instance of scrambling), or at the right edge if the antecedent follows the trace (as with extraposition).


( (IP-MAT (CONJ ac)
          (NP-NOM (PRO^N he))
          (VBD bediglode)
          (ADVP (ADV swa) (ADV +teah))
          (NP-ACC (PRO$ his) (N^A d+ada))
          (NP-DAT (D^D +tam) (N^D casere)
                  (NP-DAT-PRN (NPR^D Dioclitiane))
                  (CP-REL (WNP-NOM-1 (D^N se))
                          (C 0)
                          (IP-SUB (NP-NOM *T*-1)
                                  (BEDI w+as)
                                  (NP-NOM-PRD (NP-GEN (N^G deofles))
                                              (N^N biggencga)))))
          (. .)) (ID AelfLives,+ALS_[Sebastian]:8.7))

( (IP-MAT (CONJ and)
          (NP-NOM *con*)
          (NP-ACC (D^A +ta)
                  (CP-REL *ICH*-1))     <--- right-edge trace from below
          (NP-DAT (NPR^D Gode))
          (VBD gebrohte)
          (CP-REL-1 (WNP-2 0) 
                    (C +te)
                    (IP-SUB (NP *T*-2)
                            (NP-NOM (D^N se) (N^N deofol))
                            (VB +atbredan)
                            (MDD wolde)))
          (. .)) (ID AelfLives,+ALS_[Sebastian]:21.17))

( (IP-MAT (CONJ and)
          (NP-NOM *con*)
          (BEDI wear+d)
          (NP-DAT-1 (PRO^D him))
          (ADVP-TMP (ADV^T +ta))
          (PP (NP-DAT *ICH*-1)          <--- left-edge trace from above
              (P tomiddes))
          (. FULL-STOP)) (ID AelfLives,+ALS_[Sebastian]:50.32))

Although the elements of finite clauses are usually contiguous (but see Extraction from finite subordinate clauses), the elements of non-finite clauses (e.g., small clauses and infinitives) are often more mobile, especially in the poetry. In general in these cases, the verb or predicate of the clause is taken as representing the position of the clause, and all separated constituents are traced to this position. But see also Tracing in small clauses for some exceptions to this rule.


( (IP-MAT (ADVP-TMP (ADV^T +da))
          (NP-NOM (PRO^N ic))
          (IP-INF (VB l+adan)
                  (NP-DAT-ADT *ICH*-1)
                  (NP-ACC *ICH*-2)
                  (PP *ICH*-3)
                  (NP-ACC-SBJ *ICH*-4))
          (VBDI gefr+agn)
          (NP-DAT-ADT-1 (NP-GEN (N^G leoda))
                        (N^D weorode))
          (NP-ACC-2 (ADJ^A leofne) (N^A lareow))
          (PP-3 (P to)
                (NP-DAT (NP-GEN (N^G lides))
                        (N^D stefnan)))
          (, ,)
          (NP-ACC-4 (N^A m+acgas) (ADJ^A modgeomre))
          (. .))
  (ID coandrea,50.1706.744))

Empty quantifier heads

In some cases a quantifier head taking a PP complement is understood. In these cases, in order to make the structure clear, an empty Q is inserted.


(NODE (IP-MAT (PP (P +After)
                  (NP-DAT (D^D +tysum)))
              (RP+AXDI ongunnon)
              (NP-NOM (Q *)
                      (PP (P of)
                          (NP-DAT (D^D +dam) (N^D gegaderwyrhtum))))
              (VB t+alan)
              (NP-ACC (D^A +done) (N^A halgan)
                      (CP-REL (WNP-NOM-1 0)
                              (C +te)
                              (IP-SUB (NP-NOM *T*-1)
                                      (VBD geh+alde)
                                      (NP-ACC (D^A +tone) (N^A preost)))))
              (. ,))
      (ID AelfLives,+ALS_[Maur]:186.396)) 

( (IP-MAT (NEG+CONJ ne)
          (ADVP-TMP (ADV^T +tonne))
          (NP (Q *)
              (PP (P nemne)
                  (NP-ACC (NP-ACC (ADJ^A medmicel)
                                  (N^A d+al)
                                  (NP-GEN (N^G hlafes)))
                          (CONJP (CONJ &)
                                 (NP-ACC (NUM^A an)
                                         (NP-GEN (N^G henne))
                                         (N^A +ag)
                                         (PP (P mid)
                                             (NP-ACC (Q^A litle) (N^A meolc))
                                             (PTP-ACC (NP-DAT (N^D w+atre))
                                                      (VBN^A $gemengede))))))
))
          (NP-NOM (PRO^N he))
          (RP+VBDI onfeng)
          (. .))
      (ID Bede3,Bede_3:17.230.30.912)) 

(NODE (IP-SUB-SPE (NP-NOM (PRO^N hit))
                  (ADVP (NEG+ADV na))
                  (NEG+BEDI n+as)
                  (QP-NOM-PRD (Q *)
                              (PP (P buton)
                                  (NP-DAT (N^D hete) (CONJ &) (N^D gewinnum))
)))
      (ID Orosius,Or_3:11.77.10.1431))

See also Focus particles for the NE...BUTON construction.

Direct Speech

All IPs and CPs that are part of direct speech sequences have a final label -SPE. Non-clausal speech is labelled QTP.


( (IP-MAT-SPE (CONJ ac) 
              (NP-NOM *pro*)
              (VBPS secge)
              (CP-FRL-SPE (WNP-ACC-1 (D^A +t+at))
			  (C 0)
			  (IP-SUB-SPE (NP-ACC *T*-1)
				      (NP-NOM (PRO^N he))
				      (VBPS secge)))
              (, ,)
              (IP-MAT-PRN (NP-NOM (PRO^N he))
                          (VBDI cw+a+d))
              (. ,)) (ID WHom,WHom_2:39.27))

( (CODE )
  (IP-MAT (CONJ And)
          (NP-NOM (D^N +t+at) (N^N godspel))
          (VBDI cw+a+d)
          (, :)
          (QTP (NP-NOM (N^N Wa))
               (NP-DAT (D^D +dam) (N^D wifum)
                       (CP-REL-SPE (WNP-NOM-1 0)
				   (C +te)
				   (IP-SUB-SPE (IP-SUB-SPE (NP-NOM *T*-1)
							   (ADVP-TMP (ADV^T +tonne))
							   (VBPI tyma+d))
					       (CONJP (CONJ &)
						      (IP-CON-SPE (NP-NOM *T*-1)
								  (PP (P on)
								      (NP-DAT (D^D +tam) (ADJ^D earmlican) (N^D timan)))
								  (NP-ACC (PRO$ heora) (N^A cild))
								  (VBPI feda+d))))))
               (. .))) (ID WHom,WHom_5:12.171))

Internal Structure of clauses

Internal structure of matrix clauses (IP-MAT)
Imperatives UTON clauses Parenthetical matrix clauses (IP-MAT-PRN) Glosses (+T+AT IS...) Parenthetical constructions with verbs of naming
Internal structure of subordinate clauses
Extraction from finite subordinate clauses Clausal subjects
PPCME2 Table of contents
YCOE Table of contents

Internal structure of matrix clauses (IP-MAT)

PPCME2

Imperatives

*difference*
Imperatives are labelled IP-MAT in the YCOE (not IP-IMP as in the PPCME2). This is because of the potential ambiguity between some subjunctives and imperatives. Imperative clauses only have a subject if one is present. Empty subjects are not added in imperatives.


( (CODE )
  (IP-MAT-SPE (VBI Ar+ara+d)
              (NP-ACC (PRO$^A eower) (N^A sigebecn))
              (PP (P fram)
                  (NP-DAT (ADJ^D eor+dlicum) (N^D gewilnungum)))
              (. ,)) (ID AelfLives,+ALS_[Sebastian]:59.36))

( (IP-MAT-SPE (CONJ ac)
              (VBI gelyf)
              (PP (P on)
                  (NP-ACC (D^A +tone) (N^A H+alend)))
              (. FULL-STOP)) (ID AelfLives,+ALS_[Sebastian]:187.110))

( (CODE )
  (IP-MAT-SPE (VBPH Do)
              (IP-SMC-SPE (NP-ACC-SBJ (PRO$^A +tin) (N^A mod))
			  (ADJP-ACC-PRD (ADJ^A hluttor)))
              (CP-ADV-SPE (C +t+at)
			  (IP-SUB-SPE (NP-NOM (PRO^N +tu))
				      (VB leornian)
				      (MDPS m+age)
				      (PP (P +turh)
					  (NP-ACC (ADJ^A so+df+ast) (N^A gescead)))
				      (CP-QUE-SPE (WNP-NOM-1 (WPRO^N hwa))
						  (C 0)
						  (IP-SUB-SPE (NP-NOM-PRD *T*-1)
							      (NP-NOM (PRO$^N +tin) (N^N scyppend))
							      (BEPS sy)))))
              (. ;)) (ID AelfLives,+ALS_[Sebastian]:216.131))

UTON clauses

Clauses with finite verb UTON are labelled IP-MAT, but like in imperatives, when an overt subject is not present, no empty subject is added.


( (IP-MAT (CONJ ac)
          (UTP uton)
          (ADVP-TMP (ADV^T nu))
          (VB clypian)
          (NP-ACC (NPR^A Crist))
          (NP-DAT-ADT (PRO^D us))
          (PP (P to)
              (NP (N gefylstan)))
          (. ,)) (ID AelfLives,+ALS[Forty_Soldiers]:80.1306))

Parenthetical matrix clauses (IP-MAT-PRN)

Parenthetical matrix clauses are labelled IP-MAT-PRN. Without equal-sign coindexing they are full clauses embedded within another clause or other constituent, as for instance, asides, quotatives (CW+A+T HE), or glosses (THAT IS, ...). With an equal-sign index they are generally a way of dealing with VP conjunction of certain types or right-node raising (see Conjunction of embedded IPs and elision).


( (CODE )
  (IP-MAT (ADVP-TMP (ADV^T +Ta))
          (VBDI cw+a+d)
          (NP-NOM (NP-GEN (PRO^G heora))
                  (NUM^N an)
                  (, ,)
                  (IP-MAT-PRN (NP-NOM (PRO$ his) (N^N nama))  <--- aside
                              (BEDI w+as)
                              (NP-NOM-PRD (NPR^N Quirion))))
          (, ,)
          (IP-MAT-SPE (INTJ Eala)
                      (NP-NOM-VOC (PRO^N ge) (N^N gebro+dra))
                      (, ,) 
                      (UTP uton)
                      (BE beon)
                      (VBN^N gehyrte))
          (. ;)) (ID AelfLives,+ALS[Forty_Soldiers]:67.1295))


( (CODE )
  (IP-MAT-SPE (NEG Ne)
              (VBP awurpe)
              (NP-NOM (PRO^N ge))
              (IP-MAT-PRN (NP-NOM (PRO^N ic))   <--- quotative
                          (VBP bidde))
              (NP-ACC (PRO$^A eowerne) (ADJ^A beorhtan) (N^A sige))
              (PP (PP (P for)
                      (NP-DAT (NP-GEN (N^G wifa))
                              (N^D sw+asnyssum)))
                  (CONJP (CONJ o+d+de)
                         (PP (P for)
                             (NP-DAT (NP-GEN (N^G cyldra))
                                     (N^D tearum)))))
              (. .)) (ID AelfLives,+ALS_[Sebastian]:57.35))

Glosses (+T+AT IS...)

Glosses or other explanations introduced by +T+AT IS are common in the texts. It is difficult, however, to always distinguish them with certainty from relative clauses. YCOE policy (as in the PPCME2) is to restrict these literally to the +T+AT IS cases. The following similar types are done as relative clauses if preceded by non-final punctuation (i.e, a comma or colon), and as separate tokens if preceded by final punctuation (period or semi-colon):

the verb is past tense (+T+AT W+AS) or plural (+T+AT SYND)
the main verb is not BE (+T+AT IS GEREHT/GECWEDEN/etc.)
there is a PP indicating language (+T+AT IS IN ENGLISC/LEDEN/etc.)

Note that in these cases, the relative marker is always +T+AT, and it is treated as the relative pronoun, regardless of the number/gender of the antecedent.

Some +T+AT IS cases may be relatives, however, and it's not entirely clear that some of the other variants do not also introduce glosses. The gloss itself is contained within an XP since it can be of any category (see X Phrases). For inflected categories, the case of the gloss either matches that of the glossed element or is nominative. As usual, it is marked when it is unambiguous, otherwise not. In this case, however, the ambiguity is usually nom/acc.


(NP-NOM (PRO$ his) (N^N wissigend)
	(, ,)
	(IP-MAT-PRN (NP-NOM (D^N +t+at))
		    (BEPI is)
		    (XP (NP-NOM (N^N cr+atwisa)))))

(NP-NOM-PRD (NPR^N Golgotha)
	    (, ,)
	    (IP-MAT-PRN (NP-NOM (D^N +t+at))
			(BEPI is)
			(, ,)
			(XP (NP-NOM (NP-GEN (N^G heafodpannan))
				    (N^N stow)))))

(NP-DAT-ADT (VBN^D besmitenum) (N^D handum)
	    (, ,)
	    (IP-MAT-PRN (NP-NOM (D^N +t+at))
			(BEPI is)
			(, ,)
			(XP (NP-DAT (ADJ^D un+twogenum) (N^D handum)))))

(NODE (NP-ACC (NP-GEN (D^G +t+as) (NPR^G Appollonies))
              (N^A swurd)
              (, ,)
              (CP-REL (WNP-NOM-1 (D^N +t+at)) <--- relative with past tense verb 
                      (C 0)
                      (IP-SUB (NP-NOM *T*-1)
                              (BEDI w+as)
                              (NP-NOM-PRD (ADJ^N m+arlic) (N^N w+apn)))))
      (ID AelfLives,+ALS_[Maccabees]:295.3211)) 

(NODE (NP-NOM (NUM^N +treo) (N^N endebyrdnysse)
              (PTP-NOM (PP (P on)
                           (NP (N annysse)))
                       (VBN^N gesette))
              (, ,)
              (CP-REL (WNP-NOM-2 (D^N +t+at)) <--- relative with plural verb
                      (C 0)
                      (IP-SUB (NP-NOM *T*-2)
                              (BEPI synd)
                              (NP-PRD (FW laboratores) (, ,) (FW oratores) (, ,) (FW bellatores)))))
      (ID AelfLives,+ALS_[Maccabees]:812.3545)) 

(NODE (NP-NOM-PRD (NPR^N Emmanuhel)
                  (CP-REL (WNP-NOM-1 +t+at)   <--- relative with GEREHT
                          (C 0)
                          (IP-SUB (NP-NOM *T*-1)
                                  (BEPI is)
                                  (VBN gereht)
                                  (XP-PRD (IP-MAT (NP-NOM (NPR^N God))
                                                  (BEPI is)
                                                  (PP (P mid)
                                                      (NP (PRO us))))))))
      (ID CathHomII,+ACHom_II,_1:7.141.132)) 

(NODE (NP (FW Vigilantius)
          (, .)
          (CP-REL (WNP-NOM-2 (D^N +t+at))     <--- relative with ON ENGLISC
                  (C 0)
                  (IP-SUB (NP-NOM *T*-2)
                          (BEPI is)
                          (PP (P on)
                              (NP-ACC (ADJ^A englisc)))
                          (XP-PRD (ADJP (ADJ wacolre))))))
      (ID CathHomII,+ACHom_II,_9:72.17.1507)) 

( (CODE )
  (IP-MAT (NP-NOM (PRO^N He))
          (VBDI cw+a+d)
          (, :)
          (QTP (FW Surget) (FW gens) (FW contra) (FW gentem) (, ,) 
               (FW et) (FW reliqua))
          (. .)) (ID WHom,WHom_3:23.64))
( (CODE )
  (IP-MAT (NP-NOM (D^N +D+at))  <--- new token after final punctuation
          (BEPI is)
          (PP (P on)
              (NP-ACC (ADJ^A Englisc)))
          (, ,)
          (IP-MAT (RP upp)
                  (VBPI r+asa+d) 
                  (NP-NOM (N^N +teoda))
                  (, ,)
                  (IP-MAT-PRN (NP-NOM (PRO^N he))
                              (VBDI cw+a+d)))
          (. ,)) (ID WHom,WHom_3:23.65))

Parenthetical constructions with verbs of naming

In constructions like one of his brothers -- was called John -- went away the clause set off by dashes here is treated as a parenthetical clause with a *pro* subject.


( (CODE )
  (IP-MAT (CONJ And)
          (NP-NOM (NUM^N an)
                  (NP-GEN (PRO$ his) (N^G geferena))
                  (, ,)
                  (IP-MAT-PRN (NP-NOM *pro*) 
                              (NP-NOM-PRD (NPR^N Eleazarus))
                              (VBD hatte)))
          (, ,)
          (VBDI arn)
          (PP (P to)
              (NP-DAT (NUM^D anum) (N^D ylpe)
                      (CP-REL (WNP-NOM-2 0)
                              (C +te)
                              (IP-SUB (NP-NOM *T*-2)
                                      (ADVP-LOC (ADV^L +d+ar))
                                      (ADJP-NOM-PRD (ADJ^N enlicost))
                                      (BEDI w+as)))))
          (. ,)))

(CONJP (CONJ and)
       (IP-MAT=0 (NP-NOM *con*) 
           (PP (P be)
               (NP-DAT (Q^D sumum) (N^D cynincge)
                    (IP-MAT-PRN *ICH*-2)
                    (NP-NOM-PRN *ICH*-3)))
           (NP (PRO eow))
           (VB cy+dan)
           (ADVP-TMP (ADV^T git))
           (, ,)
           (IP-MAT-PRN-2 (NP-NOM-1 *pro*) 
                      (IP-SMC (NP-NOM *-1)
                           (NP-NOM-PRD (NPR^N Abgarus)))
                      (BEDI w+as)
                      (VBN geciged))
           (, ,)
           (NP-NOM-PRN-3 (Q^N sum) (ADJ^N ges+alig) (N^N cynincg)
                      (PP (P on)
                       (NP-DAT (NP (NPR Syrian))
                            (N^D lande))))))

Internal structure of subordinate clauses

Extraction from finite subordinate clauses
Clausal subjects
Restarts

PPCME2

Extraction from finite subordinate clauses

Extraction from finite subordinate clauses often, but not always, occurs in the context of CP recursion. Generally only one element is extracted although in some cases more than one element (so far only in Bede). In all cases the extracted elements are put in the specCP position, although in some cases this is not a particularly convincing linguistic analysis. All extracted elements are traced back into the IP-SUB.


( (IP-MAT (CONJ and)
          (NP-NOM *con*)
          (VBDI het)
          (IP-INF (NP-ACC (PRO^A hine))
                  (VB warnian)
                  (, ,)
                  (CP-THT (PP-2 (P gif)
                                (CP-ADV (C 0)
                                        (IP-SUB (NP-NOM (PRO^N he))
                                                (MDD wolde)
                                                (VB libban))))
                          (, ,)
                          (C +t+at)
                          (IP-SUB (PP *ICH*-2)
                                  (NP-NOM (PRO^N he))
                                  (NEG+BEDS n+are)
                                  (PP (P on)
                                      (NP-DAT (D^D +dam) (N^D mynstre)))
                                  (ADVP-TMP (NEG+ADV^T n+afre))
                                  (ADVP-TMP (ADV^T eft))
                                  (VBN gesewen)
                                  (PP (P of)
                                      (NP-DAT (D^D +dam)
                                              (ADJ^D andweardan)
                                              (N^D d+age)
                                              (CP-REL (WADVP-TMP-1 0)
                                                      (C +te)
                                                      (IP-SUB (ADVP-TMP *T*-1)
                                                              (NP-NOM (PRO^N he))
                                                              (PP (P of)
                                                                  (NP-DAT (N^D dea+de)))
                                                              (VBDI aras))))))))
          (. .))
  (ID AelfLives,+ALS_[Maur]:206.410))

( (CODE )
  (IP-MAT (CONJ Ac)
          (NP-NOM (PRO^N he))
          (ADVP (ADV so+dlice))
          (VBD besceawode)
          (, ,)
          (CP-THT (C +t+at)                         <--- CP recursion
                  (CP-THT (NP-NOM-LFD-2 (D^N +da)
                                        (CP-REL (WNP-NOM-1 0)
                                                (C +de)
                                                (IP-SUB (NP-NOM *T*-1)
                                                        (PP (P mid)
                                                            (NP-DAT (ADJ^D so+dre) (N^D arf+astnysse)))
                                                        (PP (P on)
                                                            (NP-DAT (N^D dea+te)))
                                                        (VBPI geendia+d))))
                          (, ,)
                          (C +t+at)
                          (IP-SUB (NP-NOM-LFD *ICH*-2)
                                  (NP-NOM-RSP (PRO^N hi))
                                  (PP (P mid)
                                      (NP-DAT (NPR^D Drihtne)))
                                  (HVPI habba+d)
                                  (NP-ACC (D^A +ta) (ADJ^A selestan) (N^A gife))
                                  (PP (P on)
                                      (NP-DAT (D^D +tam) (ADJ^D so+dan) (N^D life))))))
          (. .))
  (ID AelfLives,+ALS_[Maccabees]:476.3340))

Clausal subjects

In situ clausal subjects are labelled -SBJ. Extraposed clausal subjects are co-indexed with an overt or empty expletive. (No that-clause subjects yet)


( (IP-MAT-SPE (IP-INF-SBJ-SPE-3 (TO To)
				(VB^D sittanne)
				(PP (PP (P on)
					(NP-ACC (PRO$^A mine) (ADJ^A swi+tran) (N^A healfe)))
				    (CONJP (CONJ o+d+de)
					   (PP (P on)
					       (NP-ACC (ADJ^A wynstran))))))
              (NEG+BEPI nys)
              (NP-DAT (NP-DAT (PRO^D me))
                      (CONJP *ICH*-1))
              (IP-INF-SPE (IP-INF-SPE *-3)
			  (NP (PRO inc))
			  (TO to)
			  (VB^D syllanne))
              (CONJP-1 (CONJ ac)
                       (NP-DAT (D^D +tam)
                               (CP-REL-SPE (WNP-2 0)
					   (C +te)
					   (IP-SUB-SPE (NP *T*-2)
						       (NP-NOM (PRO^N hyt))
						       (PP (P fram)
							   (NP-DAT (PRO$^D minum) (N^D F+ader)))
						       (VBN gegearwod)
						       (BEPI ys)))))
              (. .)) (ID WSGospels,Mt_[WSCp]:20.23.1341))

Restarts

Restarts are indicated with equal-sign coindexing like cases of elision (although linguistically the two are not of the same type). The most common restart type is to repeat a complementizer and subject pronoun, but more complicated cases, including those on +TA clauses also occur.


(NODE (IP-SUB (NP-NOM (PRO^N hi))
              (NP-DAT-RFL-ADT (PRO^D him))
              (PP (P to)
                  (NP (N gewunan)))
              (VBDI genaman)
              (, ,)
              (CP-THT-0 (C +t+at)
                        (IP-SUB (NP-NOM (PRO^N heo))
                                (PP (P +turh)
                                    (NP-ACC (Q^A eall)
                                            (N^A ger)
                                            (PP (P buton)
                                                (NP (NUM fiftig)
                                                    (NP-GEN (N^G neahta)
                                                            (PP (P ofer)
                                                                (NP (NPR Eastron))))))))
                                (, ,)
                                (CP-THT-PRN=0 (C +t+at)   <--- restart
                                              (NP-NOM (PRO^N heo)))
                                (NP-INS-TMP (NP-INS (D^I +ty) (ADJ^I feor+dan) (N^I wicd+age))
                                            (CONJP (CONJ &)
                                                   (NP-INS (D^I +ty) (ADJ^I syxtan))))
                                (VBPS f+aston)
                                (PP (P to)
                                    (NP-GEN (N^G nones))))))
      (ID Bede3,Bede_3:3.162.8.102)) 

(NODE (IP-MAT (VBD Wilnade)
              (NP-NOM (PRO^N he))
              (, ,)
              (CP-THT-0 (C +t+atte)
                        (IP-SUB (NP-NOM (PRO^N he))
                                (CP-FRL-LOC (WADVP-LOC-1 (ADV swa) (WADV^L hw+ar))
                                            (C swa)
                                            (IP-SUB (ADVP-LOC *T*-1)
                                                    (NP-NOM (PRO^N he))
                                                    (NP-ACC (ADJ^A gelimplice) (N^A stowe))
                                                    (VB findan)
                                                    (MDD meahte)))
                                (, ,)
                                (CP-THT-PRN=0 (C +t+at)    <--- restart
                                              (NP-NOM (PRO^N he)))
                                (MDD wolde)
                                (PP (P for)
                                    (NP (NP-GEN (NPR^G Godes))
                                        (N noman)))
                                (PP (P in)
                                    (NP (N el+teodignisse)))
                                (VB lifian)))
              (. .))
      (ID Bede3,Bede_3:14.210.6.669)) 

(IP-MAT (ADVP-TMP (ADV^T Eft))
	(NP-NOM (D^N +ta) (NPR^N Gotan))
	(ADVP-LOC (ADV^L +t+ar))
	(NP-ACC-TMP (Q^A l+assan) (N^A hwile))
	(VBDI hergedan)
	(CP-ADV-0 (C +t+at)
		  (IP-SUB (IP-SUB (NP-NOM (PRO^N hie))
				  (PP (PP (P for)
					  (NP-DAT (NP-GEN (D^G +t+as) (N^G cristendomes))
						  (N^D are)))
				      (CONJP (CONJ &)
					     (PP (P +turh)
						 (NP (NP-GEN (NPR^G Godes))
						     (N ege)))))
				  (CP-ADV-PRN=0 (C +t+at)    <--- restart
					        (NP-NOM (PRO^N hie)))
				  (NEG+CONJ na+ter)
				  (NEG+CONJ ne)
				  (NP-ACC (D^A +ta) (N^A burg))
				  (NEG ne)
				  (VBDI b+arndon))
			  (CONJP (NEG+CONJ ne)
				 (...)))))

(IP-MAT (PP (P +After) (D^D +t+am)
	    (CP-ADV (C +te)
		    (IP-SUB (NP-NOM (NPR^N Romeburg))
			    (VBN getimbred)
			    (BEDI w+as)))
	    (NP-DAT-EXT (NUMP-DAT (NUM V) (NUM^D hunde)
				  (CONJP *ICH*-1))
			(N^D wintrum)
			(CONJP-1 (CONJ &) (NUM^D feowertegum))))
	(, ,)
	(PP-0 (ADVP-TMP (ADV^T +ta))
	      (P +ta)
	      (CP-ADV-0 (C 0)
			(IP-SUB (NP-NOM (NP-NOM (NPR^N Lucius) (NPR^N Amilius))
					(CONJP (CONJ &)
					       (NP-NOM (NPR^N Paulus) (NPR^N Publius))
					       )
					(CONJP (CONJ &)
					       (NP-NOM (NPR^N Terrentius) (NPR^N Uarra))))
				(, ,)
				(PP-PRN=0 (P +ta) (PRO^N hie))   <--- restart of +TA clause 
				(BEDI w+aron)
				(NP-NOM-PRD (N^N consulas)))))
	(, ,)
	(NP-NOM (PRO^N hie))
	(VBDI geforan)
	(PP (P mid)
	    (NP-DAT (N^D firde)))
	(PP (P angean)
	    (NP-ACC (NPR^A Hannibal)))
	(. .))
      (ID Orosius,Or_4:9.101.1.1995)

Copular constructions

Predicates (-PRD)
Verbs which take a -PRD argument Passive verbs taking predicates Predicates in small clauses
Non-subject/predicate NPs in copular constructions
PPCME2 Table of contents
YCOE Table of contents

Predicates (-PRD)

In order to distinguish NP/QP/ADJP/NUMP acting as predicates from those in other roles, the extended label -PRD is added. Although in general non-adverbial QPs and NUMPs are labelled as NPs at IP-level, in copular constructions they are not (see Quantifier phrases for other uses of IP-level QP, and Number phrases for NUMPs). Verbal predicates, including participles, are not labelled -PRD with active verbs, but may be with a passive (see Passive verbs taking predicates).


    they were (NP-NOM-PRD the three wisest men)
    they were (ADJP-NOM-PRD wise)
    they were (NUMP-NOM-PRD three)
    they were (QP-NOM-PRD many)
    they were (VBN slain)

Non-nominative NPs are generally not predicates. Accusatives (except in small clauses) and datives never are. Genitives may be, as in the following examples.


( (IP-MAT-SPE (CONJ and)
              (NP-NOM (PRO$^N eower) (N^N godas))
              (NEG ne)
              (BEPI synd)
              (NP-GEN-PRD (NEG+Q^G nahtes))
              (. .)))

(PP (P +teah)
    (CP-ADV (C 0)
	    (IP-SUB (NP-NOM (PRO^N he))
		    (NEG ne)
		    (BEPS beo)
		    (NP-GEN-PRD (PRO$^G eowres) (N^G cynnes)))))


(CP-THT (C +t+at)
	(IP-SUB (NP-NOM (N^N eor+de))
		(BEPI is)
		(NP-GEN-PRD (NPR^G Drihtnes))))

PPs and ADVPs are never labelled as predicates following our policy of using minimal labelling with these categories because of the difficulties in distinguishing types consistently, nor are clauses, apart from free relatives.

Verbs which take a -PRD argument

The most common verb taking a predicate is of course BE, but it is also used with a few other verbs, namely:

BECUMAN, HATTE (be called), L+ATAN (regard, consider), LIFAN (remain) (+TURH)WUNIAN (remain), +TYNCAN (seem),

(NODE (IP-MAT (NP-NOM-PRD (NPR^N Saul))
              (VBD hatte)
              (NP-NOM (D^N se) (ADJ^N forma) (N^N cyning)
                      (CP-REL (WNP-NOM-1 0)
                              (C +te)
                              (IP-SUB (NP-NOM *T*-1)
                                      (PP (P ofer)
                                          (NP-ACC (NP-GEN (NPR^G Godes))
                                                  (N^A folc)))
                                      (VBD rixode))))
              (. .))
      (ID AelfLives,+ALS_[Book_of_Kings]:1.1835)) 

(NODE (IP-MAT (NP-NOM (NP-NOM (D^N +Tis) (ADJ^N unges+alige) (N^N gear))
                      (CONJP (CONJ &)
                             (NP-NOM (D^N +t+at) (ADJ^N godlease))))
              (ADVP-TMP (ADV^T gen))
              (PP (P to)
                  (NP-DAT (N^D d+age)))
              (ADJP-NOM-PRD (ADJ^N la+de))
              (VBPI wuna+d)
	      (...)))

(NODE (IP-SUB (NP-NOM-x *exp*)
              (NP-DAT (PRO^D him))
              (VBD +tuhte)
              (ADJP-NOM-PRD (ADJ^N sarlic))
              (, ,)
              (PP (P gif)
                  (CP-ADV (C 0)
                          (IP-SUB (NP-NOM (PRO^N he))
                                  (NEG ne)
                                  (VBDS gehulpe)
                                  (NP-DAT (NP-DAT (D^D +tam) (ADJ^D earman))
                                          (CONJP (CONJ &)
                                                 (NP-DAT (D^D +tam) (ADJ^D astyptan) (N^D wife))))))))
      (ID GDC1,:2.18.9.104))

Passive verbs taking predicates

In addition, some verbs are treated as taking predicates in the passive. These include:

CWEDAN, RECCAN (interpret), SECGAN, SEON, TELAN, +TYNCAN, (GE)WEOR+DAN, the verbs of naming: HATEN, CIGAN, NAMAN, and BERAN in the construction "be born X"

Although participial predicates with BE receive no additional label or phrase structure (beyond the POS tag VBN, etc.) (see Predicates), in the passive case it is necessary to distinguish the non-finite main verb from the participial predicate. For this reason, participial predicates with passives are given a participial phrase label (PTP) plus case and -PRD.


(NODE (IP-MAT (ADVP-TMP (ADV^T +Ta))
              (BEDI w+as)
              (NP-NOM (PRO^N he)
                      (NP-NOM-PRN (D^N se) (N^N dema)))
              (ADJP-NOM-PRD (ADJ^N yrre))
              (BEN geworden)
              (. ;))
      (ID Bede1,Bede_1:7.36.22.128)) 

(NODE (IP-SUB (NP-NOM (PRO^N heo))
              (NP-NOM-PRD (NPR^N Dere))
              (VBN^N nemde)
              (BEDI w+aron))
      (ID Bede2,Bede_2:1.96.26.43)) 

(NODE (IP-MAT-PRN (CONJ and)
                  (NP-NOM (PRO$ heora) (N^N modor))
                  (BEDI w+as)
                  (NP-NOM-PRD (NPR^N Martia))
                  (VBN gecyged))
      (ID AelfLives,+ALS_[Sebastian]:34.22)) 

(NODE (IP-MAT (NP-NOM (FW Cathedra))
              (BEPI is)
              (VBN gereht)
              (NP-NOM-PRD (N^N bisceopstol))
              (PP (P on)
                  (NP-ACC (ADJ^A Englisc)))
              (. ,))
      (ID AelfLives,+ALS[Peter's_Chair]:4.1051)) 

(NODE (IP-SUB-SPE (NP-NOM (D^N +tis) (N^N fyr))
                  (BEPS sy)
                  (ADJP-NOM-PRD (ADJ^N egeslic) (CONJ &) (Q^N micel))
                  (VBN gesegen))
      (ID Bede3,Bede_3:14.214.7.717)) 

(NODE (IP-SUB (NP-NOM *T*-2)
              (PP (P mid)
                  (D^I +ty)
                  (CP-ADV (C 0)
                          (IP-SUB (NP-NOM (PRO^N heo))
                                  (ADVP-TMP (ADV^T +ta))
                                  (ADVP-TMP (ADV^T gena))
                                  (PP (P in)
                                      (NP (N ungeleafsumnesse)))
                                  (BEDI w+aron))))
              (, ,)
              (NP-DAT (D^D +tissum) (ADJ^D maanfullum) (N^D gesinscipum))
              (BEDI w+aron)
              (PTP-NOM-PRD (VBN^N gemengde))    <--- participial predicate
              (VBN^N s+agde))
      (ID Bede1,Bede_1:16.70.24.488)) 

(NODE (IP-MAT (NP-NOM (NPR^N Crist))
              (BEDI w+as)
              (ADJP-NOM-PRD (NP-GEN (Q^G ealra) (N^G bearna)
                                    (CP-REL *ICH*-1))
                            (ADJ^N betst))
              (VBN geboren)
              (CP-REL-1 (WNP-NOM-2 0)
                        (C +te)
                        (IP-SUB (NP-NOM *T*-2)
                                (ADVP-TMP (ADV^T +afre))
                                (VBN geboren)
                                (BEDS wurde)))
              (. ,))
      (ID WHom,WHom_5:37.191)) 

(NODE (IP-MAT (NP-NOM (PRO$ His) (N^N word)
                      (CONJP (CONJ &)
                             (NX-NOM (N^N weorc))))
              (BEPI beo+d)
              (ADJP-NOM-PRD (ADJ^N gode))
              (VBN^N ge+duhte)
              (NP-DAT (ADJ^D unweran) (N^D mannum))
              (. ,))
      (ID WHom,WHom_9:114.762)) 

(IP-MAT (CONJ ac)
	(NP-NOM (D^N +tas))
	(BEPI synd)
	(NP-NOM-PRD (D^N +ta) (ADJ^N fyrmestan))
	(ADVP (ADV +deh))
	(PP (P +turh)
	    (NP (N h+a+denscipe)))
	(VBN^N getealde))

Predicates in small clauses

Predicates are also marked in small clauses.


(IP-MAT-SPE (VBI Do)
	    (IP-SMC-SPE (NP-ACC-SBJ (PRO$^A +tin) (N^A mod))
			(ADJP-ACC-PRD (ADJ^A hluttor))))


( (IP-MAT (CONJ and)
          (NP-NOM *con*)
          (VBDI gemetton)
          (IP-SMC (NP-ACC-SBJ (PRO^A hine))
                  (ADJP-ACC-PRD (ADJ^A untrumne)))
          (. ,)) (ID AelfLives,+ALS_[Sebastian]:256.154))

Non-subject/predicate NPs in copular constructions

It is often difficult to classify the dative (and less commonly genitive) NPs in copular constructions as arguments or adjuncts (dative of interest, etc.). In cases in which it is clear that the NP is an argument of the adjective (e.g. GELICE HIM, FUL WINES) the NP is contained within the ADJP. In all other cases the NP is put at IP-level. These NPs are never labelled - ADT since their status is unclear. In non-copular constructions the argument/adjunct distinction for dative NPs is generally clearer and datives of interest, etc. are labelled NP-DAT-ADT (see Arguments vs. Adjuncts).


(IP she is 
    (ADJP-NOM-PRD (ADJ^N gelice)
		  (NP-DAT (PRO^D him))))

(IP it is
    (ADJP-NOM-PRD (ADJ^N clear))
    (NP-DAT (PRO^D him)))

( (IP-MAT (CONJ and)
          (ADVP-TMP (ADV^T +afre))
          (BEDI w+as)
          (NP-NOM (D^N se) (N^N buteruc))
          (ADJP-NOM-PRD (ADJ^N brerdful)
                        (NP-GEN (N^G wines)))
          (. .))
  (ID AelfLives,+ALS_[Maur]:281.447))

Expletive constructions

Extraposed clausal subjects
Expletive subjects with impersonal expressions
Expletive subjects with it-clefts
Expletives in temporal expressions
IS TO WIT THAT...
PPCME2 Table of contents
YCOE Table of contents

As in the PPCME2 only (H)IT is treated as an expletive. In similar constructions with +T+AT, a related clause is treated as an appositive.

Extraposed clausal subjects

Extraposed clausal subjects (that-clauses, infinitives) or questions are coindexed to an expletive subject, which may be overt (HIT) or empty (*exp*).


(NODE (IP-MAT-SPE (NP-NOM-x *exp*)
                  (ADJP-NOM-PRD (ADJ^N Selre))
                  (NP-DAT (PRO^D us))
                  (BEPI is)
                  (IP-INF-SPE-x (IP-INF-SPE (TO to) (VB^D sweltenne))
                                (CONJP (CONJ and)
                                       (IP-INF-SPE (ADVP (ADV so+dlice))
                                                   (TO *)
                                                   (RP+VB anbidian)
                                                   (NP-GEN (D^G +t+as) (ADJ^G ecan) (N^G +aristes))
                                                   (PP (P +at)
                                                       (NP-DAT (D^D +dam) (ADJ^D +almihtigan) (NPR^D
Gode)))))))
      (ID AelfLives,+ALS_[Maccabees]:142.3086)) 

(NODE (IP-SUB (NP-NOM-x (PRO^N hit))
              (ADVP-TMP (ADV^T nu))
              (BEPI is)
              (ADJP-NOM-PRD (ADV to) (ADJ^N longsum))
              (IP-INF-x (NP-ACC (Q^A eall))
                        (TO to)
                        (VB^D gesecgenne)))
      (ID Orosius,Or_4:11.110.27.2234)) 

( (IP-MAT (NP-NOM-x *exp*)
          (ADJP-NOM-PRD (ADJ^N Toweard)) 
          (BEPI ys)
          (CP-THT-x (C +t+at)
                    (IP-SUB (NP-NOM (NPR^N Herodes))
                            (VBPI sec+d)
                            (IP-INF (NP-ACC (D^A +t+at) (N^A cild))
                                    (TO to)
                                    (VB^D forspillenne))))
          (. .)) (ID WSGospels,Mt_[WSCp]:2.13.95))

( (CODE )
  (IP-MAT (CONJ &)
          (NP-NOM-x (PRO^N hyt))
          (BEDI w+as)
          (BEN geworden)
          (PP (P +ta)
              (CP-ADV (C 0)
                      (IP-SUB (NP-NOM (D^N se) H+alynd)
                              (NP-ACC (D^A +tys))
                              (VBD geendude)
                              (PTP-NOM (NP-DAT (PRO$ hys) (NUM twelf) (N^D leorningcnihtum))
                                       (VAG^N (VAG bebeodende))))))
          (, ;)
          (CP-THT-x (C 0)
                    (IP-SUB (NP-NOM (PRO^N He))
                            (VBDI for)
                            (ADVP-DIX (ADV^DX +tanun))
                            (CP-ADV (C +t+at)
                                    (IP-SUB (NP-NOM (PRO^N he))
                                            (VBD (VBD l+arde) (CONJ &) (VBD bodude))
                                            (PP (P on)
                                                (NP-DAT (PRO$ hyra) (N^D burgum)))))))
          (. .)) (ID WSGospels,Mt_[WSCp]:11.1.648))

(NODE (IP-MAT (NP-NOM-x *exp*)
              (ADVP-TMP (ADV^T +Ta))
              (VBD +tuhte)
              (NP-DAT (D^D +dam) (N^D heahgerefan))
              (ADJP-NOM-PRD (ADJ^N huxlic))
              (PP (P on)
                  (NP-DAT (N^D mode)))
              (CP-THT-x (C +t+at)
                        (IP-SUB (NP-NOM (PRO^N heo))
                                (NP-ACC (ADJ^A o+terne))
                                (VBD tealde)
                                (PP (P toforan)
                                    (NP-DAT (PRO$ his) (N^D gebyrdum)))))
              (. .))
      (ID AelfLives,+ALS[Agnes]:74.555)) 

(NODE (IP-SUB (NP-NOM-x *exp*)
              (NEG+BEDI n+as)
              (ADJP-NOM-PRD (ADJ^N gesyne))
              (ADVP-TMP (ADV^T sy+d+dan))
              (PP (P on)
                  (NP-DAT (PRO$ his) (N^D hricge)))
              (CP-QUE-x (WADVP-LOC-1 (WADV^L hw+ar))
                        (C 0)
                        (IP-SUB (ADVP-LOC *T*-1)
                                (NP-NOM (D^N se)
                                        (N^N hofor)
                                        (CP-REL *ICH*-2))
                                (VBDS stode)
                                (CP-REL-2 (WNP-NOM-3 0)
                                          (C +te)
                                          (IP-SUB (NP-NOM *T*-3)
                                                  (NP-ACC (PRO^A hine))
                                                  (VBD gehefegode)
                                                  (PP (P+D^A o+d+t+at)))))))
      (ID AelfLives,+ALS_[Swithun]:104.2462)) 

(NODE (IP-SUB-SPE (NP-ACC-ADT *T*-2)
                  (NP-NOM-x *exp*)
                  (VBPI belymp+d)
                  (PP (P to)
                      (NP (PRO +te)))
                  (CP-QUE-SPE-x (WNP-GEN-1 (WADJ^G hwylcere) (N^G m+ag+de))
                                (C 0)
                                (IP-SUB-SPE (NP-GEN-PRD *T*-1)
                                            (NP-NOM (PRO^N ic))
                                            (BEPS sy))))
      (ID AelfLives,+ALS_[Alban]:55.2214))

When a clausal subject is in situ, it is labelled -SBJ.

Expletive subjects with impersonal expressions

An expletive subject is always inserted in impersonal constructions if there is no nominative element.


(NODE (IP-SUB (NP-NOM *exp*)
              (NP-ACC (PRO^A hine))
              (VBD lyste))
      (ID AelfLives,+ALS[Ash_Wed]:65.1517)) 

(NODE (IP-CON (NP-NOM *exp*)
              (NP-DAT (PRO^D him))
              (BEDI wear+d)
              (ADVP (ADV eac))
              (VBN gecydd)
              (PP (P be)
                  (NP (NP-GEN (NPR^G Maures))
                      (N tocyme))))
      (ID AelfLives,+ALS_[Maur]:125.357)) 

(NODE (IP-SUB (NP-NOM *exp*)
              (NP-DAT (PRO^D him))
              (QP-ADT (NEG+Q naht))
              (NEG ne)
              (VBDI speow))
      (ID AelfLives,+ALS[Lucy]:106.1018)) 

(NODE (IP-SUB (NP-NOM *exp*)
              (NP-DAT (PRO^D him))
              (ADVP (ADV swa))
              (VBN ge+tuht)
              (BEDS w+are))
      (ID AelfLives,+ALS_[Maur]:53.321)) 

( (IP-MAT (CONJ &)
          (NP-NOM *exp*)
          (NP-DAT (PRO^D him))
          (ADVP-TMP (ADV^T sona))
          (ADVP (ADV wel))
          (BEDI w+as)
          (. .))
  (ID Bede3,Bede_3:7.178.5.287))

Expletive subjects with it-clefts

When the HIT subject of an cleft is missing and empty expletive subject is added.


(NODE (IP-MAT (NP-NOM *exp*)
              (ADVP-TMP (ADV^T +Ta))
              (BEDI w+as)
              (PP (P +after)
                  (NP (NP-GEN (NPR^G +A+delberhtes))
                      (N for+dfore)))
              (, ,)
              (CP-CLF (WADVP-TMP-1 0)
                      (C 0)
                      (IP-SUB (ADVP-TMP *T*-1)
                              (NP-NOM (NPR^N Eadbald)
                                      (NP-NOM-PRN (PRO$ his) (N^N sunu)))
                              (VBDI feng)
                              (PP (P to)
                                  (NP-DAT (D^D +dam) (N^D rice)))))
              (. ,))
      (ID Bede2,Bede_2:5.110.22.182))

Expletives in temporal expressions


( (IP-MAT (NP-NOM *exp*)        "it was early night"
          (BEDI w+as)
          (NP-NOM-PRD (ADJ^N foreweard) (N^N niht))
          (. FULL-STOP))
      (ID Bede2,Bede_2:9.126.31.346)) 

(NODE (IP-MAT (NP-NOM *exp*)
              (BEDI W+as)
              (NP-INS-TMP (D^I +ty)
                          (ADJ^I +triddan)
                          (N^I d+age)
                          (NP-GEN (NPR^G Mai)
                                  (NP-GEN-PRN (D^G +t+as) (N^G mon+tes))))
              (PP (ADVP (ADV hu) (ADV hugu))
                  (P ymb)
                  (NP-ACC (D^A +ta)
                          (ADJ^A teog+dan)
                          (N^A tid)
                          (NP-GEN (N^G d+ages))))
              (. .))
      (ID Bede3,Bede_3:19.240.21.1003))

IS TO WIT THAT...

The construction IS TO WIT THAT... is treated as follows. The expletive and that-clause are co-indexed, and the subject is traced to the infinitive to indicate that is also the object (as in Passives with infinitives). Some other verbs used in this construction are ONGIETAN, GELYFAN, GEHIERAN, UNDERSTANDAN.


(NODE (IP-MAT (NP-NOM-1-x *exp*)
              (BEPI IS)
              (ADVP-TMP (ADV^T NV))
              (ADVP (ADV EAC))
              (IP-INF (NP *-1)
                      (TO to)
                      (VB^D witenne))
              (CP-THT-x (C +t+at)
                        (IP-SUB (NP-NOM (MAN^N man))
                                (VBPI witna+d)
                                (ADVP-TMP (ADV^T foroft))
                                (NP-ACC (NP-ACC (D^A +da) (ADJ^A arleasan) (N^A scea+dan))
                                        (CONJP (CONJ and)
                                               (NP-ACC (D^A +ta) (ADJ^A swicolan) (N^A +deofas))))))
              (. ,))
      (ID AelfLives,+ALS_[Alban]:155.2279))

In the similar construction with +T+AT subject instead of an expletive the that-clause is treated as appositive on the +T+AT as usual.


( (CODE )
  (IP-MAT (NP-NOM-2 (D^N +T+at)
                    (CP-THT-PRN *ICH*-1))
          (BEPI is)
          (ADVP (ADV hw+a+dre))
          (IP-INF (NP *-2)
                  (TO to)
                  (VB^D weotanne))
          (, ,)
          (CP-THT-PRN-1 (C +t+atte)
                        (IP-SUB (NP-NOM (D^N +t+at))
                                (BEPI is)
                                (PP (P on)
                                    (NP-DAT (ADJ^D gastlicum) (N^D geryne)))
                                (RP+VBN ongyten)))
          (. .)) (ID Bede1,Bede_1:16.76.9.530))

Subordinate clauses

Adverbial clauses (CP-ADV)
That-clauses (CP-THT)
Degree complements (CP-DEG)
Comparative clauses (CP-CMP, CP-CMPX)
Questions(CP-QUE)
Exclamations (CP-EXL)
Relative clauses CP-REL)
Free relative clauses (CP-FRL)
Reduced relatives (RRC)
Clefts (CP-CLF)
Infinitival clauses (IP-INF, IP-INF-NCO)
Small clauses (IP-SMC)
Absolutes (PTP-case-ABS)
PPCME2 Table of contents
YCOE Table of contents

All subordinate clauses which have a complementizer and/or wh- position (whether overt or not) are labelled CP with an extended label to indicate the type of clause. Subordinate clauses without a complementizer or wh-position position are labelled IP, again with an extended label to indicate type. All CPs except direct questions and all IPs except matrix IPs and imperatives (IP-MAT) are subordinate. IPs immediately dominated by CPs are labelled IP-SUB. Conjunct subordinate clauses are labelled IP-CON.

    IP subordinate clauses 
                            
    IP-CON          conjunct clause
    IP-INF          complement infinitive 
    IP-INF-NCO      non-comp. infinitive
    IP-SMC          small clause                                
    IP-SUB          subordinate IP          

    CP subordinate clauses

    CP-ADV       adverbial clause
    CP-CAR       clause-adjoined relative
    CP-CLF       it-cleft
    CP-CMP       comparative
    CP-DEG       degree complement
    CP-EOP       empty-operator CP
    CP-FRL       free relative
    CP-QUE       question
    CP-REL       relative clause
    CP-THT       that complement

Adverbial clauses (CP-ADV)

Bare CP-ADV
Adverbial +T+AS and +TY clauses
+TAM GELICOST +TE
GELICE &
SAM ... SAM
PPCME2 Table of contents
YCOE Table of contents

Clauses labelled CP-ADV are either the complement of a preposition (subordinating conjunction) or stand alone but function as adverbials. The clausal object of the preposition rather than the dominating prepositional phrase receives the marking as an adverbial clause but the adverbial function is, of course, the function of the PP as a whole. See Prepositional phrases for adverbial clauses headed by prepositions.

Bare CP-ADV

Bare adverbial clauses are usually headed by the complementizer +T+AT, or (+T+ATTE in Orosius) and very occasionally by +TE. Their most common meaning is SO THAT, but they can also mean IN THAT, UNTIL, BECAUSE, etc.


(NODE (IP-MAT (NP-NOM (PRO^N He))
              (VBD gesette)
              (NP-ACC (PRO$ his) (N^A tacn))
              (PP (P on)
                  (NP-DAT (PRO$^D minum) (N^D nebbe)))
              (CP-ADV (C +t+at)
                      (IP-SUB (NP-NOM (PRO^N ic))
                              (NP-ACC (NEG+Q^A n+anne) (ADJ^A o+derne))
                              (PP (P ofer)
                                  (NP-ACC (PRO^A hine)))
                              (NEG ne)
                              (VBP lufige)))
              (. .))
      (ID AelfLives,AELS[Agnes]:34.516)) 

( (IP-SUB (NP-NOM (PRO^N hi))
          (NP-ACC (N^A god))
          (VBPS don)
          (, ,)
          (CP-ADV (C +te)
                  (IP-SUB (NP-NOM (PRO^N hi))
                          (MDPI willa+d)
                          (VB gedwellan)
                          (NP-ACC (D^A +ta)
                                  (N^A gife)
                                  (NP-GEN (NP-GEN (ADJ^G o+dera) (N^G manna))
                                          (N^G weorces))))))
  (ID GDC1,GD_1_[C]:10.76.23.767)) 

( (IP-SUB (NP-NOM (D^N +ta) (N^N folc))
          (NP-DAT-1 (PRO^D him))
          (ADVP (NP-INS-EXT (D^I +ty))
                (ADV swi+tor))
          (PP (NP-DAT *ICH*-1)
              (P to))
          (VBDS buge)
          (CP-ADV (C +te)
                  (IP-SUB (NP-NOM (PRO^N he))
                          (HVD h+afde)
                          (NP-ACC (NP-GEN (PRO$ hiera) (N^G ealdhlafordes))
                                  (N^A sunu))
                          (PP (P on)
                              (NP-DAT (PRO$ his) (N^D gewealde))))))
  (ID Orosius,Or_3:11.80.30.1519))

Adverbial +T+AS and +TY clauses

Clauses headed by +T+AS and +TY that have an adverbial meaning (purpose/cause) are treated in a similar way to MID +TY, etc. clauses. No relation is indicated between the determiner and the clause. Both are labelled literally and dominated by an adjunct NP node.


(NP-GEN-ADT (D^G +t+as)
	    (CP-ADV (C +te)
		    (IP-SUB ...)))


(NP-INS-ADT (D^I +ty)
	    (CP-ADV (C +te)
		    (IP-SUB ...)))


( (IP-MAT (CONJ &)
          (NP-NOM (PRO^N hie))
          (MDD wolde)
          (VB drifan)
          (PP (P to)
              (NP-DAT (NP-GEN (D^G +t+as) (N^G cyninges))
                      (N^D tune)))
          (NP-INS-ADT (D^I +ty)
                      (CP-ADV (C 0)
                              (IP-SUB (NP-NOM (PRO^N he))
                                      (NEG+VBD nyste)
                                      (CP-QUE (WNP-NOM-1 (WPRO^N hw+at))
                                              (C 0)
                                              (IP-SUB (NP-NOM-PRD *T*-1)
                                                      (NP-NOM (PRO^N hie))
                                                      (BEDI w+aron))))))
          (. ;))
  (ID ChronA,:787.1.578)) 

(IP-SUB (NP-NOM (D^N +ta) (NPR^N Egipti))
	(NP-ACC (QP-ACC (ADV swa) (Q^A lytle))
		(N^A +toncunge))
	(VBDI wiston)
	(NP-DAT (NPR^D Iosepe))
	(NP-GEN-ADT (D^G +t+as)
		    (CP-ADV (C +te)
			    (IP-SUB (NP-NOM (PRO^N he))
				    (NP-ACC (PRO^A hi))
				    (PP (P +at)
					(NP-DAT (N^D hungre)))
				    (VBD ahredde))))
	(, ,))

( (IP-MAT (NP-NOM *con*)
          (VBPI li+t)
          (NP-DAT-ADT (PRO^D him))
          (PP (P on)
              (NP-DAT (N^D londe)))
          (NP-GEN-ADT (D^G +t+as)
                      (CP-ADV (C 0)
                              (IP-SUB (NP-NOM (PRO$ his) (N^N lufu))
                                      (VBPI b+ade+d))))
          (. .))
  (ID coexeter,160.93.554))

As with the MID +TY etc. case, when the CP-ADV is separated from the determiner, it is treated as extraposed and traced. Note that this is almost certainly not the right analysis linguistically (more likely is that these are correlative constructions with the second occurance suppressed +ty ... (+ty) +te ...), but it marks overtly the connection between the two parts of the clause.

+TAM GELICOST +TE

The phrase +TAM GELICOST +TE meaning just as though, just as if functions at sentence level as either an adjective or an adverb taking a clause. The relation between the three parts is quite unclear and thus we have given it minimal structure.


(ADJP (D^D +tam) (ADJ gelicost)
      (CP-ADV (C +te)
	      (IP-SUB ...)))

(ADVP (D^D +tam) (ADV gelicost)
      (CP-ADV (C +te)
	      (IP-SUB ...)))


(IP-SUB (NP-NOM (PRO^N hio))
	(BEPS sie)
	(ADJP-NOM-PRD (D^D +t+am) (ADJ^N gelicost)
		      (CP-ADV (C +te)
			      (IP-SUB (IP-SUB (NP-NOM (MAN^N mon))
					      (VBPS nime)
					      (NP-ACC (NUM^A +anne)
						      (NP-GEN (N^G eles))
						      (N^A dropan)))
				      (, ,)
				      (CONJP (CONJ &)
					     (IP-CON (NP-NOM *con*)
						     (VBPS drype)
						     (PP (P on)
							 (NP-ACC (NUM^A an) (Q^A micel) (N^A fyr)))))
				      (...)))))

(IP-MAT-SPE (NP-NOM (D^N +T+at))
	    (BEPI sindon)
	    (NP-NOM-PRD (D^N +ta) (ADJ^N godan) (N^N tida)
			(CP-REL-SPE (WNP-1 0)
				    (C +te)
				    (IP-SUB-SPE (NP *T*-1)
						(NP-NOM (PRO^N hie))
						(ADVP-TMP (ADV^T ealneg))
						(RP+VBPI foregielpa+d))))
	    (, ,)
	    (ADVP (ADV gelicost) (D^D +t+am)
		  (CP-ADV-SPE (C +te)
			      (IP-SUB-SPE (NP-NOM (PRO^N hie))
					  (ADVP-TMP (ADV^T nu))
					  (VBPS cwe+ten)
					  (CP-THT-SPE (C +t+at)
						      (IP-SUB-SPE ...))))))

GELICE &

The phrase GELICE &, which occurs only in Orosius (4x) in the meaning as though, clearly introduces a subordinate clause, although everything else about its structure is quite opaque. We have given it the structure of other adverbial clauses by labelling GELICE as a preposition and introducing the clause with a CP-ADV; the "&" is labelled literally as a conjunction.


(PP (P gelice) (CONJ &)
    (CP-ADV (C 0)
	    (IP-SUB ...)))


(IP-SUB (NP-ACC (PRO^A hie))
	(NP-NOM (NUM^N an) (N^N cyning))
	(ADVP (ADV swa) (ADV ie+delice))
	(PP (ADVP (ADV forneah))
	    (P buton)
	    (NP-DAT (Q^D +alcon) (N^D gewinne)))
	(PP (P on)
	    (NP-ACC (PRO$ his) (N^A geweald)))
	(VB be+tridian)
	(MDD sceolde)
	(, ,)
	(PP (P gelice) (CONJ &)
	    (CP-ADV (C 0)
		    (IP-SUB (NP-NOM (PRO^N hie))
			    (NP-DAT (PRO^D him))
			    (VAG +teowiende)
			    (BEDI w+aron))))
	(, ,))

SAM ... SAM

SAM ... SAM in the meaning whether ... or is labelled P. A following +TE if present is taken as a complementiser. The SAM GE ... SAM GE example in Clark Hall, where GE is as a conjunction, is an error; GE is a personal pronoun in this example, not a conjunction.


( (CODE )
  (IP-MAT (CONJ &)
          (PP (P +teah)
              (CP-ADV (C 0)
                      (IP-SUB (NP-NOM (MAN^N man))
                              (VBD asette)
                              (IP-SMC (NP-ACC-SBJ (NUM^A twegen) (N^A f+atels)) 
                                      (ADJP-ACC-PRD (ADJ^A full)
                                                    (NP-GEN (N^G eala+d) (CONJ o+d+de) (N^G w+ateres)))))))
          (, ,)
          (NP-NOM (PRO^N hy))
          (VBPI gedo+d) 
          (CP-THT (C +t+at)
                  (IP-SUB (NP-NOM (ADJ^N o+ter))
                          (BEPI bi+d)
                          (RP+VBN oferfroren)))
          (, ,)
          (PP (P sam) 
              (CP-ADV (C 0)
                      (IP-SUB-0 (NP-NOM (PRO^N hit))
                                (BEPS sy)
                                (NP-NOM-PRD (N^N sumor)))))
          (PP (P sam)
              (CP-ADV (C 0)
                      (IP-SUB=0 (NP-NOM-PRD (N^N winter)))))
          (. .)) (ID Orosius,Or_1:1.17.36.247))

That-clauses (CP-THT)

PPCME2

That-clauses with an empty complementizer

Although empty complementizers are rare in that-clauses in Old English, they do occasionally appear. They are particularly common in complement clauses to HIT GELAMP, and HIT GEWORDEN W+AS. The complement clause in these cases has some root properties (allows left-dislocations, shows inversion with +TA), but since overt complementizer +T+AT is also common, they are all treated as subordinate clauses.


(NODE (IP-MAT (CONJ And)
              (NP-NOM-x (PRO^N hit))
              (BEDI w+as)
              (BEN geworden)
              (CP-THT-x (C 0)
                        (IP-SUB (PP (P +ta)
                                    (CP-ADV (C 0)
                                            (IP-SUB (NP-NOM (D^N se) (N^N H+alend))
                                                    (VBD geendode)
                                                    (NP-ACC (D^A +tas) (N^A bigspel)))))
                                (, ,)
                                (ADVP-TMP (ADV^T +ta))    <--- inversion
                                (VBD ferde)
                                (NP-NOM (PRO^N he))
                                (ADVP-DIX (ADV^DX +tanone))))
              (. .))
      (ID WSGospels,Mt_[WSCp]:13.53.901)) 

(NODE (IP-MAT (NP-NOM-x (PRO^N Hit))
              (VBDI gelamp)
              (ADVP (ADV eac))
              (NP-INS-TMP (Q^I sume) (N^I d+age))
              (, ,)
              (CP-THT-x (C 0)
                        (IP-SUB (PP (ADVP-TMP (ADV^T +ta))
                                    (P +ta)
                                    (CP-ADV (C 0)
                                            (IP-SUB (NP-NOM (D^N se) (ADJ^N ylca)(ADJ^N arwyr+da)
                                                            (N^N wer)
                                                            (NP-NOM-PRN (NPR^N Benedictus)))
                                                    (BEDI w+as)
                                                    (PP (P in)
                                                        (NP-DAT (PRO$ his) (N^D mynstre))))))
                                (, ,)
                                (ADVP-TMP (ADV^T +ta))
                                (BEDI w+as)
                                (RP+VAG utgangende)
                                (NP-NOM (D^N se) (RP+VBN^N foresprecena) (N^N cniht)
                                        (NP-NOM-PRN (NPR^N Placidus))
                                        (NP-NOM-PRN (D^N se) (N^N munuc)
                                                    (NP-GEN (D^G +t+as) (ADJ^G halgan) (N^G weres))))
                                (IP-INF-NCO (TO to)
                                            (VB^D hladane)
                                            (NP-ACC (N^A w+ater))
                                            (PP (P of)
                                                (NP-DAT (D^D +t+are) (N^D ea))))))
              (. .))
      (ID GDC2,:7.114.22.255)) 

(NODE (IP-MAT (NP-NOM-x (PRO^N Hit))
              (VBDI gelamp)
              (ADVP (ADV niwan))
              (, ,)
              (CP-THT-x (C +t+at)          <--- with overt complementizer
                        (IP-SUB (PP (P on)
                                    (NP-ACC (N^A niht)))
                                (, ,)
                                (PP (P +ta)
                                    (CP-ADV (C 0)
                                            (IP-SUB (NP-NOM (D^N +ta) (N^N g
ebro+dra))
                                                    (BEDI w+aron)
                                                    (PP (P on)
                                                        (NP (N r+aste))))))
                                (, ,)
                                (ADVP-TMP (ADV^T +ta))
                                (VBDI comon)
                                (ADVP-LOC (ADV^L +t+ar))
                                (NP-NOM (D^N +ta) (NPR^N Langbearde))))
              (. ,))
      (ID GDC2,:17.140.34.592))

Degree complements (CP-DEG)

SWILC +T+AT clauses

PPCME2

Degree complements are contained in, or traced to, the constituent containing SWA in the SWA (X) +T+AT type of degree phrase. In the TO +T+AM/+TON type, the degree complement is contained in, or traced to, the NP dominating +T+AM/+TON.


(ADJP (ADV swa) (ADJ happy)
      (CP-DEG ...))


(NP (ADJP (ADV swa) (ADJ happy)
          (CP-DEG *ICH*-1))
    (N man)
    (CP-DEG-1 ...))


(NP (QP (ADV swa) (Q micclum)
        (CP-DEG ...)))


(ADVP (PP (P to)
          (NP-DAT (D^D +t+am)
                  (CP-DEG *ICH*-1)))
      (ADV swi+te)
      (CP-DEG-1 ...))

(IP ...
    (PP (P to)
        (NP-INS (N^I +ton)
                (CP-DEG *ICH*-1)))
    ...
    (CP-DEG-1 ...))

(NODE (IP-MAT (NP-NOM (PRO^N He))
              (BEDI wear+d)
              (ADVP-TMP (ADV^T +da))
              (ADJP-NOM-PRD (ADV swa) (ADJ^N halig)
                            (CP-DEG (C +t+at)
                                    (IP-SUB (NP-NOM (PRO^N he))
                                            (VBD geh+alde)
                                            (NP-ACC (ADJ^A untrume)))))
              (. ,))
      (ID AelfLives,+ALS[Agnes]:391.776)) 

( (IP-MAT (CONJ and)
          (NP-NOM *con*)
          (BEDI w+as)
          (ADVP (ADV swa)
                (CP-DEG *ICH*-1))
          (VBN geh+aled)
          (CP-DEG-1 (C +t+at)
                    (IP-SUB (PP (P on)
                                (NP-DAT (PRO$ hire) (N^D lice)))
                            (NEG+BEDI n+as)
                            (ADJP-NOM-PRD (ADJ^N gesyne))
                            (NP-NOM (Q^N aht)
                                    (NP-GEN (D^G +t+ara) (ADJ^G sarra) (N^G wunda)))))
          (. .))
  (ID AelfLives,+ALS[Agnes]:277.693)) 

(NODE (IP-MAT (NP-NOM (PRO^N He))
              (BEPI is)
              (NP-NOM-PRD (ADJP-NOM (ADV swa) (ADJ^N mihtig)
                                    (CP-DEG *ICH*-1))
                          (N^N wyrhta)
                          (CP-DEG-1 (C +t+at)
                                    (IP-SUB (NP-NOM (PRO^N he))
                                            (MDPI m+ag)
                                            (VB awendan)
                                            (NP-ACC (N^A yfel))
                                            (PP (P to)
                                                (NP-DAT (N^D gode)))
                                            (PP (P +turh)
                                                (NP (PRO$ his) (N godnysse))))))
              (. .))
      (ID AelfLives,+ALS[Forty_Soldiers]:315.1460)) 

(NODE (IP-MAT=0 (NP-NOM (QP (ADV swa) (Q fela)
                            (CP-DEG *ICH*-1)))
                (PP (P binnan)
                    (NP-DAT (NUM twelf) (N^D mon+dum)))
                (CP-DEG-1 (C +t+at)
                          (IP-SUB (NP-NOM (MAN^N man))
                                  (NP-ACC (PRO^A hi))
                                  (VB getellan)
                                  (NEG ne)
                                  (MDD mihte))))
      (ID AelfLives,+ALS_[Swithun]:149.2490)) 

( (IP-MAT (CONJ ac)
          (NP-NOM (PRO$ his) (N^N ge+dyld))
          (BEDI w+as)
          (QP-NOM-PRD (PP (P to)
                          (NP-INS (D^I +dan)
                                  (CP-DEG *ICH*-1)))
                      (Q^N micel))
          (, .)
          (CP-DEG-1 (C +t+at)
                    (IP-SUB (NP-NOM (PRO^N he))
                            (NEG+MDD nolde)
                            (NP-ACC (NEG+Q^A n+anne) (N^A dom))
                            (ADVP-TMP (ADV^T +da))
                            (VB gesetton)
                            (PP (ADVP-TMP (ADV^T +da))
                                (P +da)
                                (CP-ADV (C 0)
                                        (IP-SUB (NP-NOM (PRO^N he))
                                                (IP-INF-NCO (TO to) (VB^D +drowigenne))
                                                (VBDI com))))))
          (. ;))
  (ID CathHomII,+ACHom_II,_13:134.224.3039)) 

(NODE (IP-MAT (NP-NOM (N^N Win))
              (VBD ateorode)
              (PP (P +at)
                  (NP-DAT (D^D +dam) (N^D giftum)))
              (PP (P to)
                  (NP-INS (D^I +di)
                          (, .)
                          (CP-DEG (C +t+at)
                                  (IP-SUB (IP-SUB (NP-NOM (NPR^N Drihten))
                                                  (PP (P mid)
                                                      (NP-DAT (ADJ^D selran) (N^D wine)))
                                                  (NP-ACC (D^A +ta) (N^A gebeoras))
                                                  (VBD gegladode))
                                          (, .)
                                          (CONJP (CONJ and)
                                                 (IP-CON (NP-NOM *con*)
                                                         (NP-ACC (NP-GEN (PRO$ his) (N^G godcundnysse))
                                                                 (N^A mihta))
                                                         (PP (P mid)
                                                             (NP-DAT (D^D +tam) (N^D tacne)))
                                                         (VBD geopenode)))))))
              (. ;))
      (ID CathHomII,+ACHom_II,_4:30.45.741))

SWILC +T+AT clauses

SWILC +T+AT clauses are also done as degree clauses, since they are in many ways similar, although perhaps not identical.


( (CODE )
  (IP-MAT (NP-NOM (D^N Se) (ADJ^N ilca) (NPR^N Papirius))
          (BEDI w+as)
          (PP (P +after)
              (NP-DAT (D^D +t+am) (N^D gefeohte)))
          (PP (P mid)
              (NP-DAT (NPR^D Romanum)))
          (NP-GEN (ADJP-GEN (ADJ^G swelces)
                            (CP-DEG *ICH*-1)) 
                  (N^G domes))
          (VBN beled)
          (CP-DEG-1 (C +t+at)
                    (IP-SUB (NP-NOM (PRO^N hie))
                            (NP-ACC (PRO^A hiene))
                            (PP (P to)
                                (NP-INS (D^I +don)
                                        (CP-THT-PRN *ICH*-2)))
                            (VBN gecoren)
                            (HVDI h+afdon) 
                            (CP-THT-PRN-2 (C +t+at)
                                          (IP-SUB ...))))))

Comparative clauses (CP-CMP, CP-CMPX)

SWA X SWA and +TONNE comparatives
SWA comparatives
SWILCE comparatives
As if SWILCE comparatives Like SWILCE comparatives
Corelative comparatives
PPCME2 Table of contents
YCOE Table of contents

PPCME2

SWA X SWA and +TONNE comparatives

In general, comparatives are treated the same way as in the PPCME2, except that comparative adjectives and adverbs are not overtly marked as comparative at the POS level. YCOE makes use of two comparative clause labels CP-CMP and CP-CMPX. The first is used for straightforward comparative clauses, which includes virtually all cases of the SWA X SWA type, and the following +TONNE types:

when all that's missing from the comparative IP is a trace of the same category as the antecedent of the comparative (Mary is taller than Lucy is) (PP (P for) (D^I +ton) (CP-ADV-SPE (C 0) (IP-SUB-SPE (NP-NOM (PRO$^N eowre) (N^N brocu)) (ADVP-TMP (ADV^T nu)) (QP-NOM-PRD (Q^N l+assan) (PP *ICH*-4)) (BEPI sindon) (PP-4 (P +tonne) (CP-CMP-SPE (WQP-NOM-2 0) (C 0) (IP-SUB-SPE (QP-NOM-PRD *T*-2) (NP-NOM (NP-GEN (PRO^G heora)) (D^N +ta)) (BEDS w+are))))))) when some of the comparative IP is elided but the missing part can be (or could be) straightforwardly reconstructed from the matrix IP (Mary is taller than Lucy); in this case the IP is treated in the same way as other IPs with elision (that is, they are coindexed with equal-sign coindexing to the complete clause on which they are modelled). Note that in YCOE comparative prepositions always take a clause and not a single constituent as was allowed in the PPCME2 unless (as very rarely) the case of a single NP following SWA/+TONNE is not in the case expected under a clausal analysis with elision. ( (IP-MAT-SPE-0 (CONJ and) (NP-NOM (PRO^N +tu)) (BEPI bist) (ADJP-NOM-PRD (ADV swa) (ADJ^N hal) (PP (P swa) (CP-CMP-SPE (WADJP-NOM-1 0) (C 0) (IP-SUB-SPE=0 (ADJP-NOM *T*-1) (NP-NOM (PRO^N ic)))))) (. .)) (ID AelfLives,+ALS_[Sebastian]:187.110)) ( (IP-MAT (NP-NOM *con*) (VBD wiste) (CP-THT (C +t+at) (IP-SUB-0 (NP-NOM (D^N seo) (N^N dohtor) (, ,) (CP-REL (WNP-NOM-1 0) (C +te) (IP-SUB (NP-NOM *T*-1) (NP-ACC (NPR^A drihten)) (HVD h+afde) (VBN gecoren)))) (, ,) (ADVP (ADV hra+dor) (PP *ICH*-2)) (MDD wolde) (VB sweltan) (PP-2 (P +tonne) (CP-CMP (WADVP-3 0) (C 0) (IP-SUB=0 (ADVP *T*-3) (VB ceorlian)))))) (. .)) (ID AelfLives,AELS[Agnes]:301.705))

The CP-CMPX label is used for problematic cases of the SWA X SWA and +TONNE types; these include the following:

where the category of the trace is not clear ( (IP-MAT (CONJ &) (NP-NOM *con*) (BEDI w+aron) (ADVP (ADV swi+tor) (PP *ICH*-1)) (VAG winnende) (PP (P on) (NP-ACC (NPR^A Thebane))) (PP-1 (P +tonne) (CP-CMPX (NP-NOM (PRO^N hie)) (NP-GEN (N^G fultumes)) (HVD h+afde))) (. ,)) (ID Orosius,Or_3:1.55.18.992)) comparative subdeletion (Mary is taller than Jane is small) (CP-THT (C +t+at) (IP-SUB (NP-NOM (PRO^N hi)) (NP-DAT (PRO$ heora) (N^D feondum)) (ADVP (ADV bet) (PP *ICH*-3)) (VBDS dyde) (PP-3 (P +tonne) (CP-CMPX (ADVP (ADV wyrs)))) (PP (P mid) (NP-DAT (D^D +t+am))))) cases in which the comparative IP includes a subordinate clause of which the subordinator is elided (It is better that Mary leaves than [it is that] John starts shouting) ( (IP-MAT (CONJ ac) (NP-NOM-x *exp*) (NP-DAT (PRO^D me)) (BEPI is) (ADVP (ADV swa) (ADV +deah)) (ADJP-NOM-PRD (ADJ^N leofre) (PP *ICH*-1)) (CP-THT-x (C +t+at) (IP-SUB (NP-NOM (PRO^N ic)) (PP (P on) (NP-ACC (NP-GEN (NPR^G Godes)) (N^A handa))) (VBP befealle))) (PP-1 (P +tonne) (CP-CMPX (IP-SUB (NP-NOM (PRO^N ic)) (PP (P on) (NP-ACC (NP-GEN (N^G mannes)) (N^A handa))) (VBP befealle)))) (, ;) (PP (P for+dan) (CP-ADV (C +te) (IP-SUB (NP-NOM (PRO$ his) (N^N mildheortnyssa)) (BEPI syndon) (ADJP-NOM-PRD (ADJ^N m+anigfealde))))) (. .)) (ID AelfLives,AELS[Pr_Moses]:246.1744)) other cases where the parallel between the matrix and comparative IP is not sufficiently close to make equal sign coindexing helpful or it is generally unclear what is going on

Within a CP-CMPX clause, any clear structure which is present is annotated, but no attempt is necessarily made to connect the parts or reconstruct a complete clause; in particular no wh-operator or trace is indicated.

SWA comparatives

Comparatives headed only by SWA (or SWA SWA) are treated as follows. If the comparative clause is complete (i.e., all the core material (verbs and arguments) are present) then they are treated as straightforward adverbial relatives.


(IP-MAT-SPE (NP-NOM (PRO^N ic))
	    (HVP h+abbe)
	    (PP (P on)
		(NP-DAT (PRO$^D minum) (N^D hordcleofan)))
	    (NP-ACC (NUM^A an) (ADJ^A wundorlic) (N^A weorc))
	    (NP-DAT-ADT (PRO^D me))
	    (PP (P to)
		(NP (N gewissunge)))
	    (, ,)
	    (PP (P +after)
		(NP-DAT (NP-GEN (N^G steorrena))
			(N^D gesetnyssum)))
	    (PP (ADVP (ADV swa))
		(P swa)
		(CP-CMP-SPE (WADVP-1 0)
			    (C 0)
			    (IP-SUB-SPE (ADVP *T*-1)
					(NP-NOM (PRO^N hi))
					(VBPI standa+d)
					(PP (P on)
					    (NP-DAT (N^D heofonum))))))
	    (. .)) (ID AelfLives,+ALS_[Sebastian]:260.157))

If, on the other hand, any of the core material has been elided, then the -CMPX label is used, and no wh-operator or trace is included. This approach was adopted largely to avoid having to distinguish SWA clauses with adverbial gap from SWA clauses with nominal gaps.


( (IP-SUB (NP-NOM (D^N +det) (N^N hus))
	  (QP (Q eall))
	  (VBDI scean)
	  (, ,)
	  (PP (ADVP (ADV swa))
	      (P swa)
	      (CP-CMPX (NP-NOM (N^N sunne))
		       (PP (P on)
			   (NP-ACC (N^A d+ag)))))))

(IP-MAT-SPE (CONJ and)
	    (NP-NOM *con*)
	    (VBDI nam)
	    (NP-ACC (PRO$ his) (ADJ^A agene) (N^A swystor))
	    (, ,)
	    (PP (P to)
		(NP-DAT (PRO$ his) (ADJ^D fulum) (N^D synscype)))
	    (, ,)
	    (PP (ADVP (ADV swa))
		(P swa)
		(CP-CMPX-SPE (NP-NOM (PRO^N ge))
			     (VBPI r+ada+d)
			     (PP (P on)
				 (NP-DAT (PRO$^D eowrum) (N^D gerecednyssum)))))
	    (. .)) (ID AelfLives,+ALS_[Sebastian]:172.102))

*difference*
Note that SWA clauses are treated quite differently in the YCOE from the way AS clauses are treated in the PPCME2. Apart from SWA +T+AT purpose/result clauses and temporal SWA clauses, all SWA clauses are labelled as comparatives.

SWILCE comparatives

As if SWILCE comparatives

Most SWILCE clauses in which SWILCE is a conjunction/preposition rather than an agreeing form of SWILC are as if clauses and are done as straightforward adverbial comparatives (see SWA comparatives).


( (CODE )
  (IP-MAT (NP-NOM *con*)
          (VBD Eode)
          (ADVP-TMP (ADV^T +ta))
          (ADVP (ADV bli+delice))
          (PP (P to)
              (NP-DAT (D^D +dam) (ADJ^D blindum) (N^D cwearterne)))
          (, ,)
          (PP (P swylce) 
              (CP-CMP (WADVP-1 0) 
                      (C 0)
                      (IP-SUB (ADVP *T*-1)
                              (NP-NOM (PRO^N heo))
                              (BEDS w+are)
                              (VBN gela+dod)
                              (PP (P to)
                                  (NP-DAT (ADJ^D lustfullum) (N^D beorscype))))))
          (. ,)) (ID AelfLives,+ALS[Agatha]:97.854))

( (IP-MAT (CONJ and)
          (NP-NOM *con*)
          (VBDI cw+a+d)
          (CP-THT (C +t+at)
                  (IP-SUB (NP-NOM *exp*) 
                          (NP-DAT (PRO^D him))
                          (VBD +tuhte)
                          (PP (P swylce)
                              (CP-CMP (WADVP-1 0)
                                      (C 0)
                                      (IP-SUB (ADVP *T*-1)
                                              (NP-NOM (PRO^N he))
                                              (VBD eode)
                                              (PP (P ofer)
                                                  (NP (N blostman))))))))
          (. .)) (ID AelfLives,+ALS_[Sebastian]:379.231))

As with SWA comparatives, if any of the core material of the clause is missing in an AS IF comparative, -CMPX is used.


( (IP-MAT (CONJ and)
	  (NP-NOM (PRO^N hi) (Q^N ealle))
	  (VBDI cw+adon)
	  (PP (P swylce)
	      (CP-CMPX (PP (P mid)
			   (NP-DAT (NUM^D anum) (N^D mu+de)))))
	  (, ,)
	  (IP-MAT-SPE (VBI Do+d)
		      (ADVP-TMP (ADV^T nu))
		      (PP (P be)
			  (NP (PRO us)))
		      (NP-ACC (D^A +t+at)
			      (CP-REL-SPE (WNP-ACC-1 (D^A +t+at))
					  (C 0)
					  (IP-SUB-SPE (NP-ACC *T*-1)
						      (NP-NOM (NPR^N Drihten))
						      (MDP wile)))))
	  (. .)) (ID AelfLives,+ALS[Forty_Soldiers]:131.1342))

Like SWILCE comparatives

SWILCE clauses which are best translated with like or of the sort which can be either predicates or dependent on an antecedent NP. Those which are dependent on an antecedent NP are contained within that NP (or traced to it) and have a WNP operator and subject trace. Like other SWILCE/SWA clauses, they are labelled -CMPX if incomplete.


( (CODE )
  (IP-MAT (NP-NOM (PRO^N He))
          (VBD worhte)
          (NP-ACC (NUM^A twa) (ADJ^A st+anene) (N^A tabulan)
                  (PP (P swylce) 
                      (CP-CMPX (NP-NOM (D^N +ta) (ADJ^N o+tre))
                               (BEDI w+aron))))
          (. ,)) (ID Heptateuch,Exod:34.4.3573))

(IP-MAT (NP-NOM (PRO$ Hys) (N^N ansyn))
	(BEDI w+as)
	(PP (P swylce) 
	    (CP-CMPX (NP-NOM (N^N ligyt)))))

(IP-SUB-SPE (NEG+BEPI nis)
	    (NP-NOM (NEG+Q^N nan) (ADJ^N o+ter) (ADJ^N swilc) (NPR^N God))
	    (PP (P swilce) 
		(CP-CMPX-SPE (NP-NOM (PRO$^N ure) (NPR^N God)))))

( (CODE )
  (IP-MAT (NP-NOM (D^N +Ta)
		  (PP *ICH*-2))
          (VBDI foron)
          (PP (P ofer)
              (NP-ACC (Q^A eall) (NPR Egypta) (N^A land)))
	  (, ,)
	  (PP-2 (P swilce)
		(CP-CMP (WNP-NOM-1 0)  
			(C 0)
			(IP-SUB (IP-SUB (NP-NOM *T*-1)
					(ADVP-TMP (NEG+ADV^T n+afre))
					(PP (P +ar)
					    (NP-DAT (D^D +d+are) (N^D tide)))
					(NEG+BEDI n+aron))
				(, ,)
				(CONJP (NEG+CONJ ne)
				       (IP-CON (NP-NOM *T*-1)
					       (ADVP-TMP (ADV^T +after))
					       (ADJP-NOM-PRD (ADJ^N towearde))
					       (NEG ne)
					       (BEPI synd))))))
          (. .)) (ID Heptateuch,Exod:10.14.2780))

( (IP-MAT-SPE (ADVP (ADV Witodlice))
              (ADVP-TMP (ADV^T nu))
              (PP (P to)
                  (NP-ACC (N^A morgen)))
              (PP (P on)
                  (NP-DAT (D^D +tysse) (ADJ^D ylcan) (N^D tide)))
              (NP-NOM (PRO^N ic))
              (VBP sende)
              (NP-ACC (Q^A micelne) (N^A hagol)
                      (, ,)
                      (PP (P swylce) 
                          (CP-CMP-SPE (WNP-NOM-1 0) 
				      (C 0)
				      (IP-SUB-SPE (NP-NOM *T*-1)
						  (PP (P on)
						      (NP-DAT (NPR Egypta) (N^D lande)))
						  (NEG+BEDI n+as)
						  (PP (P fram)
						      (NP-DAT (D^D +dam) (N^D d+age)
							      (CP-REL-SPE (WADVP-TMP-2 0)
									  (C +te)
									  (IP-SUB-SPE (ADVP-TMP *T*-2)
										      (NP-NOM (PRO^N hit))
										      (VBN gesta+tolod)
										      (BEDI w+as)))))
						  (PP (P o+d)
						      (NP-ACC (D^A +das) (N^A dagas)))))))
              (. .)) (ID Heptateuch,Exod:9.17.2719))

Corelative comparatives


( (CODE )
  (IP-MAT (CONJ Ac)
          (QP-ADT (QP-INS-EXT (ADV swa) (Q^I myccle))
                  (Q ma))
          (NP-NOM (PRO^N he))
          (VBD ar+afnede) 
          (PP (P +after)
              (NP-INS (D^I +ton)))
          (NP-ACC (ADJ^A hefigran) (N^A gewin))
          (, ,)
          (QP-INS-ADT (ADV swa) (Q^I myccle) 
		      (PP (P swa)
			  (CP-CMP (WQP-1 0) 
				  (C 0)
				  (IP-SUB (NP-NOM (PRO^N he))
					  (QP-ADT (QP *T*-1)
						  (Q ma))
					  (VBD (VBD gemette) (CONJ &) (RP+VBDI onfand))
					  (NP-ACC (D^A +tone) (N^A lareow) 
						  (ADJP-ACC (ADJ^A selfne))
						  (NP-GEN (D^G +t+as) (N^G bealwes))
						  (PTP-ACC (ADVP (ADV openlice))
							   (VAG^A feohtendne)
							   (PP (P ong+an)
							       (NP-ACC (PRO^A hine)))))))))
          (. .)) (ID GDC2,:8.121.11.350))

( (IP-MAT (CONJ &)
          (NP-NOM (PRO^N he))
          (MDPI m+ag)
          (ADVP (QP-INS-EXT (ADV swa) (Q^I mycele))
                (ADV gewislicor))
          (VB secgan)
          (NP-ACC (Q^A gehwylc) (N^A +ting))
          (PP (P be)
              (NP-DAT (PRO^D him)))
          (, ,)
          (QP-ADT (QP-INS-EXT (ADV swa) (Q^I mycle)) 
                  (Q ma)
                  (CP-CMP (WQP-1 0)
                          (C 0)
                          (IP-SUB (QP-ADT *T*-1)
                                  (NP-NOM-x *exp*) 
                                  (NP-DAT (PRO^D him))
                                  (VBDI gelamp)
                                  (, ,)
                                  (CP-THT-x (C +t+at)
                                            (IP-SUB (NP-NOM (PRO^N he))
                                                    (MDD sceolde)
                                                    (PP (P betwyh)
                                                        (NP-DAT (PRO$ his) (N^D +tegnungum)))
                                                    (BE beon))))))
          (. .)) (ID GDC1,:9.56.23.569))

( (IP-MAT (CONJ &)
          (NP-NOM *con*)
          (QP-ADT (QP-INS-EXT (ADV swa) (Q^I myccle))
                  (Q ma))
          (RP+AXDI ongan)
          (NP (PRO$ his) (N scylde))
          (VB (VB weopan) (CONJ &) (VB scamian))
          (, ,)
          (QP-DAT-ADT (ADV swa) (Q^D mycclum) 
                      (PP (P swa)
                          (CP-CMP (WQP-2 0)
                                  (C 0)
                                  (IP-SUB (NP-NOM (PRO^N he))
                                          (NP-ACC-RFL (PRO^A hine) 
                                                      (ADJP-ACC (ADJ^A sylfne)))
                                          (QP-ADT (QP *T*-2)
                                                  (Q ma))
                                          (RP+VBDI ong+at)
                                          (NP-ACC (ADJ^A +afweardne) (N^A agyltan))
                                          (PP (P beforan)
                                              (NP-DAT (NP-GEN (PRO$ his) (N^G f+ader)
                                                              (NP-GEN-PRN *ICH*-1))
                                                      (N^D eagum)
                                                      (NP-GEN-PRN-1 (NPR^G Benedictes))))))))
          (. .)) (ID GDC2,:13.129.34.470))

( (IP-MAT (CONJ &)
          (QP-ADT (QP-INS-EXT (ADV swa) (Q^I mycele)) 
                  (Q ma)
                  (CP-CMP (WNP-1 0)
                          (C 0)
                          (IP-SUB (NP-ADT *T*-1)
                                  (NP-NOM (PRO^N he))
                                  (VBPI gefealh)
                                  (PP (P mid)
                                      (NP (N geornnysse)))
                                  (NP-DAT (D^D +tam) (N^D gebedum)))))
          (, ,)
          (QP-ADT (QP-INS-EXT (ADV swa) (Q^I myccle))
                  (Q ma))
          (NP-NOM (PRO^N he))
          (RP+VBDI ong+at)
          (IP-INF (PP (NP-DAT (PRO^D him))
                      (P ong+an))
                  (VB standan)
                  (PP (P in)
                      (NP-DAT (NUM^D anum) (N^D lichaman)))
                  (NP-NOM (D^N +t+at) (N^N weorod)
                          (NP-GEN (NP-GEN (D^G +tara) (N^G deofla)) 
                                  (N^G m+anigeo))))
          (. .)) (ID GDC1,:10.74.14.769))

Questions (CP-QUE)

Whether questions
WHETHER as WPRO or WQ

PPCME2

Whether questions

Matrix WHETHER questions often (but not always) have the word order of indirect questions, even when they are clearly direct. Although word order is a major clue for distinguishing direct from indirect questions, there are usually enough other indications (especially in edited texts) to allow this distinction to be made not taking word order into account. Like other questions, then, direct WHETHER questions do not contain a (C 0) position even when they do not exhibit subject-verb inversion. If the C position is filled, however, it is necessarily present. This will in some cases result in direct questions with uninverted word order.


( (CODE )
  (CP-QUE-SPE (WQ Hw+a+der)      <--- direct whether question with inversion
              (IP-SUB-SPE (BEDI w+as)
			  (NP-NOM (NP-GEN (NPR^G Iohannes))
				  (N^N fulluht))
			  (PP (CONJ +te)
			      (PP (P of)
				  (NP-DAT (N^D heofonum)))
			      (CONJP (CONJ +te)
				     (PP (P of)
					 (NP-DAT (N^D mannum))))))
              (. ;)) (ID WSGospels,Mt_[WSCp]:21.25.1416))


(CP-QUE-SPE (WQ hw+a+ter)        <--- direct whether question without inversion
	    (IP-SUB-SPE (NP-NOM (D^N +ta) (N^N ealdras))
			(RP+VBPS ongyton)
			(CP-THT-SPE (C +t+at)
				    (IP-SUB-SPE (NP-NOM (D^N +tis))
						(BEPI is)
						(NP-NOM-PRD (NPR^N Crist))))))

( (CP-QUE-SPE (WQ Hw+a+ter)  <--- direct whether question with overt C
              (C +te)
              (IP-SUB-SPE (NP-NOM (PRO$^N +tin) (N^N eage))
			  (ADJP-NOM-PRD (ADJ^N manful))
			  (BEPI ys)
			  (PP (P for+tam)
			      (CP-ADV-SPE (C +te)
					  (IP-SUB-SPE (NP-NOM (PRO^N ic))
						      (ADJP-NOM-PRD (ADJ^N god))
						      (BEPI eom)))))
              (. ?)) (ID WSGospels,Mt_[WSCp]:20.15.1318))

( (IP-MAT-SPE (UTP utun)
              (VB geseon)
              (CP-QUE-SPE (WQ hw+a+ter) <--- indirect whether question
                          (C 0)
                          (IP-SUB-SPE (IP-SUB-SPE (NP-NOM (NPR^N Helias))
						  (VBPS cume))
				      (CONJP (CONJ &)
					     (IP-CON-SPE (NP-NOM *con*)
							 (MDP wylle)
							 (NP-ACC (PRO^A hyne))
							 (VB alysan)))))
              (. .)) (ID WSGospels,Mt_[WSCp]:27.49.2066))

WHETHER as WPRO or WQ

WHETHER is tagged WPRO only when appropriately inflected; if the appropriate inflection is zero, then it is tagged WPRO only when there is a gap/trace in its following clause. Otherwise, WHETHER is labelled WQ. WHETHER will always be WQ when the alternatives are specified, joined by +TE...+TE.


(CP-QUE-SPE (WNP-ACC-1 (WPRO^A hw+a+terne)   <--- inflected HW+A+TER
		       (PP *ICH*-2))
	    (IP-SUB-SPE (MDP wylle)
			(NP-NOM (PRO^N ge))
			(CP-THT-SPE (C +t+at)
				    (IP-SUB-SPE (NP-ACC *T*-1)
						(NP-NOM (PRO^N ic))
						(VBP forgyfe)
						(NP (PRO eow))))
			(PP-2 (P of)
			      (NP-DAT (D^D +tisum) (NUM^D twam)))))

Exclamations (CP-EXL)

PPCME2

Relative clauses (CP-REL)

SE +TE relatives
+T+AS clauses Temporal +T+AS clauses According as +T+AS clauses
+T+AT relatives
Temporal and locative relatives
PPCME2 Table of contents
YCOE Table of contents

PPCME2

SE +TE relatives

When a SE (+TE) relative has a possible antecedent NP, the SE is taken as the relative pronoun and +TE, if present, as a complementizer. When +TE is not present an empty complementizer is added. This is done invariably although these cases are in fact ambiguous, the other alternative being that the SE +TE clause is an appositive NP on the antecedent.


(NP antecedent
    (CP-REL (WNP se)
	    (C +te)
	    (IP-SUB ...)))

(NP antecedent
    (CP-REL (WNP se)
	    (C 0)
	    (IP-SUB ...)))

( (CODE )
  (IP-MAT (ADVP-TMP (ADV^L Her))
          (NP-NOM (NPR^N Landfranc)
                  (CP-REL (WNP-NOM-1 (D^N se))
                          (C +te)
                          (IP-SUB (NP-NOM *T*-1)
                                  (BEDI w+as)
                                  (NP-NOM-PRD (N^N abbod))
                                  (PP (P an)
                                      (NP-DAT (NPR^D Kadum)))))
                  (CP-REL *ICH*-2))
          (VBDI com)
          (PP (P to)
              (NP-DAT (NPR^D +anglalande)))
          (, ,)
          (CP-REL-2 (WNP-NOM-3 (D^N se))
                    (C 0)
                    (IP-SUB (NP-NOM *T*-3)
                            (PP (P efter)
                                (NP-DAT (Q^D feawum) (N^D dagum)))
                            (BEDI wear+d)
                            (NP-NOM-PRD (N^N arcebiscop))
                            (PP (P on)
                                (NP-DAT (NPR^D Kantwareberig)))))
          (. .)) (ID ChronA,ChronA_[Plummer]:1070.1.1461))

When there is no antecedent, SE is taken as the antecedent and the relative pronoun is an empty wh-operator (WNP, etc.).


(NP (D se)
    (CP-REL (WNP 0)
	    (C +te)
	    (IP-SUB ...)))

( (IP-MAT-SPE (NP-NOM-LFD (D^N Se)
                          (CP-REL-SPE (WNP-NOM-1 0)
				      (C +te)
				      (IP-SUB-SPE (NP-NOM *T*-1)
						  (RP+VBPI ofslih+d))))
              (NP-NOM-RSP (D^N se))
              (BEPI by+t)
              (ADJP-NOM-PRD (NP-DAT (N^D dome))
                            (ADJ^N scyldig))
              (. .)) (ID GospMt1,Mt_[WSCp]:5.21.232))

In EALL +T+AT..., EALLE +TA... plus relative, EAL(LE) is taken as the antecedent and +T+AT/+TA as the relative pronoun, although some of these cases (those with an overt complementizer) are in fact ambiguous between this reading and one in which the determiner is part of the antecedent.


(NODE (IP-CON-0 (NP-NOM *con*)
                (MDPI m+ag)
                (VB don)
                (NP-ACC (Q^A eall)
                        (CP-REL (WNP-ACC-1 (D^A +t+at))
                                (C 0)
                                (IP-SUB=0 (NP-ACC *T*-1)
                                          (NP-NOM (PRO^N he))
                                          (MDP wile)))))
      (ID AelfLives,+ALS_[Alban]:8.2294)) 

(NP-NOM (Q^N ealle)
        (CP-REL-SPE (WNP-NOM-1 (D^N +da))
		    (C +te)
		    (IP-SUB-SPE (NP-NOM *T*-1)
				(PP (P on)
				    (NP-ACC (NPR^A God)))
				(VBDI truwodon))))

In deciding whether a clause beginning with a determiner is a relative or a new matrix clause with the determiner as an argument, the editor's punctuation is followed; that is, if final punctuation (a period or semicolon) precedes the determiner, it is taken as an argument of a new matrix clause, if not, the determiner is taken as heading a relative clause.

+T+AS clauses

Temporal +T+AS clauses

Temporal +T+AS (+TE) clauses, like other SE +TE clauses, may either be relatives on a temporal antecedent, in which case +T+AS is the relative pronoun, or +T+AS itself may be the antecedent, in which case the relative pronoun is an empty operator. When +T+AS is the antecedent, the operator is WADVP-TMP, as in other temporal relatives.


(NP-GEN-TMP (D^G +t+as)            <--- temporal with antecedent +T+AS
	    (CP-REL (WADVP-TMP-1 0)
		    (C +te)
		    (IP-SUB (ADVP-TMP *T*-1)
			    (...))))

(NP antecedent                     <--- relative pronoun +T+AS
    (CP-REL (WNP-GEN-1 (D^G +t+as))
	    (C +te)
	    (IP-SUB (NP-GEN-TMP *T*-1)
		    (...))))


( (IP-MAT (CONJ &)
          (NP-GEN-TMP (D^G +t+as)
                      (CP-REL (WADVP-TMP-1 0)
                              (C +de)
                              (IP-SUB (ADVP-TMP *T*-1)
                                      (NP-NOM (PRO^N hi))
                                      (QP-INS-ADT (Q^I hwon))
                                      (VBN^N gereorde)
                                      (BEDI w+aron))))
          (, ,)
          (VBDI aras)
          (NP-NOM (PRO^N he))
          (ADVP (ADV hra+de))
          (. ,))
  (ID Bede3,Bede_3:3.162.4.100)) 

(NODE (IP-MAT (NP-GEN-TMP (ADVP-TMP (ADV^T Sona))
                          (D^G +t+as)
                          (CP-REL (WADVP-TMP-1 0)
                                  (C +te)
                                  (IP-SUB (ADVP-TMP *T*-1)
                                          (NP-NOM (PRO^N heo))
                                          (VAG gehalgad)
                                          (BEDI w+as))))
              (, ,)
              (ADVP-TMP (ADV^T +ta))
              (VBD dyde)
              (NP-NOM (MAN^N mon))
              (NP (PRO$ his) (N lichoman))
              (RP in)
              (. ;))
      (ID Bede2,Bede_2:3.104.32.131)) 

( (IP-MAT (CONJ and)
          (NP-DAT-1 (PRO^D him))
          (VBDI l+ag)
          (PP (NP-DAT *ICH*-1)
              (P onuppan))
          (NP-NOM (Q fela)
                  (NP-GEN (N^G byr+dena)
                          (NP-GEN (N^G eor+dan))))
          (PP (P binnon)
              (NP-DAT (NUM seofon) (N^D nihton)
                      (, ,)
                      (CP-REL (WNP-GEN-2 (D^G +t+as))
                              (C +de)
                              (IP-SUB (NP-GEN-TMP *T*-2)
                                      (NP-NOM (PRO^N he))
                                      (VBDI forsoc)
                                      (NP-ACC (D^A +ta) (Q^A feawa) (N^A axan)))
			      )))
          (. .))
  (ID AelfLives,+ALS[Ash_Wed]:56.1513)) 

(NODE (IP-MAT (ADVP-TMP (ADV^T +Da))
              (PP (P on)
                  (NP-DAT (D^D +dam) (ADJ^D fifteogo+dan) (N^D d+age)
                          (CP-REL (WNP-GEN-1 (D^G +d+as))
                                  (C +de)
                                  (IP-SUB (NP-GEN-TMP *T*-1)
                                          (NP-NOM (PRO^N hi))
                                          (PP (P fram)
                                              (NP-DAT (NPR Egypta) (N^D lande)))
                                          (VBDI ferdon)))))
              (, .)
              (BEDI wear+d)
              (NP-NOM (NP-GEN (NPR^G Godes))
                      (N^N wuldor))
              (VBN gesewen)
              (, .)
              (PP (P on)
                  (NP-DAT (D^D +dam) (N^D westene)))
              (PP (P uppon)
                  (NP-DAT (NUM^D anum)
                          (N^D munte)
                          (CP-REL (WNP-NOM-2 (D^N se))
                                  (C 0)
                                  (IP-SUB (NP-NOM *T*-2)
                                          (BEPI is)
                                          (VBN gehaten)
                                          (NP-NOM-PRD (NPR^N Synay))))))
              (. .))
      (ID CathHomII,+ACHom_II,_12.1:113.122.2525))

According as +T+AS clauses

+T+AS clauses which mean according as are treated as relatives. +T+AS clauses with a more adverbial meaning (purpous/cause), as well as adverbial +TY clauses, are treated as adverbial clauses, since there is no plausible gap.


(NP-GEN-ADT (D^G +t+as)             <--- +T+AS = according as  
	    (CP-REL (WNP-1 0)
		    (C +te)
		    (IP-SUB (NP *T*-1)
			    (...))))

( (CODE )
  (IP-MAT (NP-NOM (NPR^N Petrus))
          (VBDI cw+a+d)
          (, :)
          (IP-MAT-SPE (NP-GEN-ADT (D^G +t+as)
                                  (CP-REL-SPE (WNP-1 0)
					      (C +te)
					      (IP-SUB-SPE (NP *T*-1)
							  (NP-NOM (PRO^N ic))
							  (RP+VBP ongyte))))
                      (, ,)
                      (NP-NOM (D^N +tes))
                      (BEDI w+as)
                      (NP-NOM-PRD (Q^N mycel) (N^N wer))
                      (ADVP-LOC (ADV^L utan))
                      (PP (P on)
                          (NP-DAT (D^D +tam) (N^D m+agnum)
                                  (CP-REL-SPE (WNP-2 0)
					      (C +te)
					      (IP-SUB-SPE (NP *T*-2)
							  (NP-NOM (PRO^N he))
							  (VBD worhte))))))
          (. ,)) (ID GDC1,GD_1_[C]:5.47.14.444))

+T+AT relatives

Relative clauses headed by +T+AT alone are potentially ambiguous between a reading with +T+AT as the relative pronoun, and one in which +T+AT is the complementizer. YCOE policy is to label +T+AT as the relative pronoun only when it matches the number/gender of the antecedent, except in non-restrictive glossing relatives, which are invariably headed by +T+AT (see Glosses (+T+AT IS) for examples), where it is always treated as the relative pronoun. Temporal relative clauses also quite commonly have complementizer +T+AT.


(NP-ACC (D^A +t+at)                     <--- neuter
	(CP-REL (WNP-ACC-2 (D^A +t+at)) <--- +T+AT relative pronoun
		(C 0)
		(IP-SUB (NP-ACC *T*-2)
			(NP-NOM (PRO^N ge))
			(ADVP-TMP (ADV^T nu))
			(VBPI witon))))

(NP-ACC (ADJ^A o+tre) (N^A flocrade)  <--- feminine 
	(, ,)
	(CP-REL (WNP-NOM-1 0)
		(C +t+at)             <--- +T+AT complementizer
		(IP-SUB (NP-NOM *T*-1)
			(VBDI rad)
			(RP ut)
			(PP (P wi+d)
			    (NP-GEN (NPR^G Lygtunes))))))

(IP-MAT-SPE (NP-NOM (D^N +t+at))
	    (BEDI w+aron)
	    (NP-NOM-PRD (ADJP-NOM (ADV swi+de) (ADJ^N swearte)) <--- plural
			(N^N men)
			(, ,)
			(CP-REL-SPE (WNP-NOM-2 0)
				    (C +t+at)
				    (IP-SUB-SPE (NP-NOM *T*-2)
						(NP (PRO me))
						(VBDI l+addon))))
	    (ID GDC1,:12.89.21.976)) 

(NP-ACC (PRO$^A minne) (N^A engel)  <--- masculine
	(, ,)
	(CP-REL-SPE (WNP-NOM-1 0)
		    (C +t+at)
		    (IP-SUB-SPE (NP-NOM *T*-1)
				(NP (PRO +te))
				(VBPS l+ade)
				(PP (RP in)
				    (P to)
				    (NP-DAT (D^D +t+are) (N^D stowe)
					    (...))))))

Temporal and locative relatives

Temporal relatives like the day she came generally do not have an overt relative pronoun and, as usual we supply an empty wh-operator (WADVP-TMP). Locative relatives often have an overt +T+AR relative pronoun, but when they don't, an empty locative wh- (WADVP-LOC) is likewise supplied. Note that +T+AR +T+AR clauses are done like SE +TE relatives with the first +T+AR acting as antecedent and the second as the relative pronoun. See also Temporal +T+AS clauses and Free relatives.


( (CODE )
  (IP-MAT (NP-NOM (D^N Se))
          (VBD awogode)
          (NP-ACC (NPR^A Constantian)
                  (NP-ACC-PRN (NP-GEN (D^G +t+as) (N^G caseres))
                              (N^A dohtor)))
          (PP (P on)
              (NP-DAT (D^D +t+are) (N^D tide)
                      (CP-REL (WADVP-TMP-1 0)     <--- temporal relative; empty wh-
                              (C +te)
                              (IP-SUB (ADVP-TMP *T*-1)
                                      (NP-NOM (D^N +da) (ADJ^N Scy+discan))
                                      (ADVP (ADV swi+de))
                                      (VBDI wunnon)
                                      (PP (P wi+d)
                                          (NP-ACC (PRO^A hine)))))))
          (. .)) (ID AelfLives,AELS[Agnes]:299.703))

( (IP-MAT (ADVP-TMP (ADV^T nu))
          (BEPI is)
          (NP-NOM (D^N se) (N^N tima)
                  (CP-REL (WADVP-TMP-1 0)
                          (C +t+at)
                          (IP-SUB (ADVP-TMP *T*-1)
                                  (NP-NOM (D^N +deos) (N^N woruld))
                                  (BEPI is)
                                  (VBN gem+ancged)
                                  (PP (PP (P mid)
                                          (NP-DAT (ADJ^D m+anigfealdan) (N^D mane)))
                                      (CONJP (CONJ &)
                                             (PP (P mid)
                                                 (NP-DAT (ADJ^D felafealdan) (N^D facne))))))))
          (. ,)) (ID WHom,WHom_5:23.180))

( (CODE )
  (IP-MAT (ADVP-TMP (ADV^T +Ta))
          (VBD asende)
          (NP-NOM (D^N se) (N^N cynincg))
          (ADVP-TMP (ADV^T eft))
          (NP-ACC (ADJ^A o+dre) (NUM fiftig))
          (PP (RP up) (P to)
              (NP (D +d+are) (N dune)
                  (CP-REL (WADVP-LOC-1 (ADV^L +t+ar))  <--- locative relative; overt wh-
                          (C 0)
                          (IP-SUB (ADVP-LOC *T*-1)
                                  (NP-NOM (NPR^N Helias))
                                  (VBDI s+at)))))
          (. ,)) (ID AelfLives,+ALS_[Book_of_Kings]:251.2035))

( (CODE )
  (IP-MAT (NP-NOM *con*) 
          (VBDI Worhton)
          (ADVP-TMP (ADV^T +ta))
          (NP-ACC (NUM^A anne) (N^A gangtun))
          (, ,)
          (ADVP-LOC (ADV^L +t+ar)                      <--- +T+AR +T+AR relative
                    (CP-REL (WADVP-LOC-1 (ADV^L +d+ar))
                            (C 0)
                            (IP-SUB (ADVP-LOC *T*-1)
                                    (NP-NOM (D^N se) (N^N god)
                                            (NP-NOM-PRN (NPR^N Baal)))
                                    (ADVP-TMP (ADV^T +ar))
                                    (BEDI w+as)
                                    (VBN gewur+dod)
                                    (ADVP (ADV wolice))
                                    (PP (P+D^A o+d+t+at)))))
          (. .)) (ID AelfLives,+ALS_[Book_of_Kings]:379.2123))

( (IP-MAT (CONJ and)
          (ADVP-LOC (ADV^L +d+ar)
                    (CP-REL *ICH*-1))
          (BEPI is)
          (ADVP-TMP (ADV^T nu))
          (NP-NOM (ADJ^N ful) (N^N w+ater))
          (CP-REL-1 (WADVP-LOC-2 (ADV^L +d+ar))
                    (C 0) 
                    (IP-SUB (ADVP-LOC *T*-2)
                            (NP-NOM (D^N +da) (ADJ^N fulan))
                            (VBDI wunodon)))
          (. ,)) (ID AelfLives,+ALS[Pr_Moses]:211.1759))

(NODE (IP-INF (NP-ACC-SBJ (PRO^A hi))
              (VB bidan)
              (PP (P on)
                  (NP-DAT (D^D +t+am) (N^D ealonde)
                          (, ,)
                          (CP-REL (WADVP-LOC-2 0)    <--- locative relative; empty wh-
                                  (C +te)
                                  (IP-SUB (ADVP-LOC *T*-2)
                                          (NP-NOM (PRO^N hi))
                                          (RP upp)
                                          (VBDI comon))))))
      (ID Bede1,Bede_1:14.58.8.368)) 

( (IP-MAT (CONJ &)
          (NP-NOM *con*)
          (PP (P mid)
              (NP-DAT (D^D +tissum)))
          (NP-ACC (D^A +ta) (N^A burg))
          (PP (P mid)
              (NP-DAT (Q^D micelre) (ADJ^D heannisse)))
          (RP+VBD ymbsealde)
          (PP (P from)
              (NP-DAT (D^D +t+am) (N^D d+ale)
                      (, ,)
                      (CP-REL (WADVP-LOC-1 0)
                              (C +te)
                              (IP-SUB (ADVP-LOC *T*-1)
                                      (NP-NOM (PRO^N heo))
                                      (NP-DAT (N^D londe))
                                      (VBN ge+teoded)
                                      (BEPI is)))))
          (. .))
  (ID Bede3,Bede_3:14.202.2.597))

Free relative clauses (CP-FRL)

Free relatives headed by a determiner
WH- Free relatives
Free relatives headed by +TE
PPCME2 Table of contents
YCOE Table of contents

PPCME2

At IP-level, nominal free relatives acting as non-subject arguments are labelled CP-FRL; that is, they have no extended label (except possibly -SPE, which can be added to any clause). Free relatives with other functions are labelled in the usual way, as follows.


subject             CP-FRL-SBJ
left-dislocation    CP-FRL-LFD
locative            CP-FRL-LOC
temporal            CP-FRL-TMP
directional         CP-FRL-DIX
other adjunct       CP-FRL-ADT
predicate           CP-FRL-PRD

As in the PPCME2, the default for ambiguous question/free relative cases is to parse them as questions.

*difference*
Free relatives never have empty phrase labels surrounding them to indicate function, as they do in the PPCME2.


( (IP-MAT-SPE (NP-NOM (PRO^N ic))
              (NP-DAT (PRO^D hire))
              (VBD sende)
              (CP-FRL-SPE (WNP-ACC-1 (D^A +t+at)) <--- argument
			  (C 0)
			  (IP-SUB-SPE (NP-ACC *T*-1)
				      (NP-NOM (PRO^N ic))
				      (NP-DAT (PRO^D hyre))
				      (VBDI behet)))
              (. ,)) (ID Heptateuch,Gen:38.23.1522))


( (IP-MAT (CONJ &)
          (NP-NOM *con*)
          (VBD beferde)
          (NP-ACC (D^A +d+at) (ADJ^A Israhelisce) (N^A folc))
          (CP-FRL-LOC (WADVP-LOC-1 (ADV^L +d+ar))  <--- locative
                      (C 0)
                      (IP-SUB (ADVP-LOC *T*-1)
                              (NP-NOM (PRO^N hi))
                              (VBN^N gewicode)
                              (BEDI w+aron)
                              (PP (P be)
                                  (NP-DAT (D^D +d+are) (ADJ^D Readan) (N^D S+a))
				  )))
          (. .)) (ID Heptateuch,Exod:14.7.2926))


( (CODE )
  (IP-MAT (CONJ And)
          (ADJP-NOM-PRD (ADJ^N so+d))
          (BEPI is)
          (CP-FRL-SBJ (WNP-ACC-1 (D^A +t+at))  <--- subject
                      (C 0)
                      (IP-SUB (NP-ACC *T*-1)
                              (NP-NOM (PRO^N ic))
                              (VBP secge)))
          (. ,)) (ID WHom,WHom_9:143.775))

( (CODE )
  (IP-MAT-SPE (NP-NOM (D^N +D+at))
              (BEPI is)
              (CP-FRL-PRD-SPE (WNP-ACC-1 (D^N +t+at))  <--- predicate
			      (C 0)
			      (IP-SUB-SPE (NP-ACC *T*-1)
					  (NP-NOM (PRO^N ic))
					  (NP (PRO eow))
					  (ADVP-TMP (ADV^T +ar))
					  (VBD s+ade)))
              (, ,)
              (IP-MAT-PRN (VBDI cw+a+t)
                          (NP-NOM (PRO^N he)))
              (. ,)) (ID Heptateuch,Gen:42.14.1738))

Free relatives headed by a determiner

Relative clauses headed by a determiner with no other possible antecedent and no complementizer are treated as free relatives. The determiner is taken as being within the clause (filling the wh-position) and the clause has no antecedent. This includes the adverbial type headed by +T+AR, +TIDER, etc. (Determiner + complementizer relatives (the SE +TE type) are treated as regular relative clauses with the determiner acting as antecedent.)


(CP-FRL (WNP-NOM-1 (D^N +t+at))
	(C 0)
	(IP-SUB (NP-NOM *T*-1)
		...))

( (IP-MAT-SPE (CONJ &)
              (CP-FRL-SBJ-SPE (WNP-NOM-1 (D^N +d+at))
			      (C 0)
			      (IP-SUB-SPE (NP-NOM *T*-1)
					  (ADVP-LOC (ADV^L +t+ar))
					  (ADJP-NOM-PRD (ADJ^N dead))
					  (BEPI bi+d)))
              (BEPI bi+t)
              (NP-GEN-PRD (PRO^G his))
              (. .)) (ID Heptateuch,Exod:21.33.3247))

( (IP-MAT-SPE (CONJ &)
              (VBI gehyr)
              (NP-ACC (PRO$ his) (N^A stemne))
              (, ,)
              (PP (P for) (D^D +tam)
                  (CP-ADV-SPE (C +te)
			      (IP-SUB-SPE (IP-SUB-SPE (NP-NOM (PRO^N he))
						      (VBPI forgyf+d)
						      (CP-FRL-SPE (WNP-ACC-1 (D^A +tone))
								  (C 0)
								  (IP-SUB-SPE (NP-ACC *T*-1)
									      (NP-NOM (PRO^N ge))
									      (VBPI syngia+d))))
					  (, ,)
					  (CONJP (CONJ &)
						 (IP-CON-SPE (NP-NOM (PRO$^N min) (N^N nama))
							     (BEPI is)
							     (PP (P on)
								 (NP-DAT (PRO^D him))))))))
              (. .)) (ID Heptateuch,Exod:23.21.3324))

( (CODE )
  (IP-MAT (CONJ &)
          (NP-NOM *con*)
          (VBD ferde)
          (ADVP-TMP (ADV^T nihtes))
          (PP (P to)
              (CP-FRL (WADVP-LOC-1 (ADV^L +t+ar))   <--- +T+AR
                      (C 0) 
                      (IP-SUB (ADVP-LOC *T*-1)
                              (NP-NOM (PRO$ heora) (N^N $fynd))
                              (VBDI wicodon))))
          (. ,)) (ID Heptateuch,Judg:7.19.4853))

( (CODE )
  (IP-MAT (CONJ &)
          (NP-GEN-TMP (ADVP (ADV ra+de))
                      (D^G +t+as)
                      (CP-REL (WADVP-TMP-1 0)
                              (C +te)
                              (IP-SUB (ADVP-TMP *T*-1)
                                      (NP-NOM (D^N +ta) (N^N senatus))
                                      (VBDI gehierdon)
                                      (CP-THT (C +t+at)
                                              (IP-SUB (NP-NOM (NPR^N Marius))
                                                      (PP (P $to)
                                                          (NP (NPR Rome)))
                                                      (VBD neal+acte))))))
          (, ,)
          (NP-NOM (PRO^N hie) (Q^N ealle))
          (RP ut)
          (VBDI aflugon)
          (PP (P on)
              (NP-ACC (NP-GEN (NPR^G Creca))
                      (N^A lond)))
          (PP (PP (P +after)
                  (NP (NPR Sillan)))
              (CONJP (CONJ &)
                     (PP (P +after)
                         (NP-DAT (NPR^D Pompeiuse)))))
          (, ,)
          (CP-FRL-DIX (WADVP-DIX-2 (ADV^DX +tider))  <--- +TIDER
                      (C 0)
                      (IP-SUB (ADVP-DIX *T*-2)
                              (NP-NOM (PRO^N hi))
                              (ADVP-TMP (ADV^T +ta))
                              (PP (P mid)
                                  (NP-DAT (N^D firde)))
                              (VBN gefaren)
                              (BEDI w+aron)))
          (. .)) (ID Orosius,Or_5:11.125.5.2539))

WH- Free relatives

WH- free relatives are mostly of the SWA WH- SWA type. SWA WH- is in the wh-position and the second SWA is in C.


(CP-FRL (WNP-ACC (ADV swa) (WPRO^A hw+at))
	(C swa)
	(IP-SUB ...))

( (IP-MAT-SPE (CONJ &)
              (VBI nym)
              (CP-FRL-SPE (WNP-ACC-1 (ADV swa) (WPRO^A hw+at))
			  (C swa)
			  (IP-SUB-SPE (NP-ACC *T*-1)
				      (NP-NOM (PRO^N +tu))
				      (NP-GEN (PRO$^G +dines))
				      (VBPS finde)))
              (. ;)) (ID Heptateuch,Gen:31.32.1262))

( (CODE )
  (IP-MAT-SPE (CP-FRL-LFD-SPE (WNP-NOM-1 (ADV Swa) (WPRO^N hwa))
			      (C swa)
			      (IP-SUB-SPE (NP-NOM *T*-1)
					  (VBPI agyt)
					  (NP-ACC (NP-GEN (D^G +d+as) (N^G mannes))
						  (N^A blod))))
              (, ,)
              (NP-NOM-RSP (PRO$ his) (N^N blod))
              (BEPI by+d)
              (VBN agoten)
              (. ;)) (ID Heptateuch,Gen:9.6.377))

( (IP-MAT-SPE (CONJ &)
              (CP-FRL-SPE (WADVP-DIX-1 (ADV swa) (WADV^DX hwyder)) 
			  (C swa)
			  (IP-SUB-SPE (ADVP-DIX *T*-1)
				      (NP-NOM (PRO^N +du))
				      (VBPI f+arst)))
              (, :)
              (VBI gemun)
              (CP-THT-SPE (C +d+at)
			  (IP-SUB-SPE (NP-NOM (PRO^N +tu))
				      (VBN gel+aht)
				      (BEDS w+are)))
              (. .)) (ID Heptateuch,Gen:20.16.855))

Other types include those headed by LOCAHW- , and a WH- alone.


( (IP-MAT (CONJ &)
          (NP-NOM (PRO^N heo))
          (VBDI cw+a+d)
          (ADVP-TMP (ADV^T eft))
          (, :)
          (IP-MAT-SPE (NP-NOM (PRO^N Ic))
                      (VBP +dolige)
                      (CP-FRL-SPE (WNP-ACC-1 (WPRO^A locahw+at))  <--- LOCAHW-
				  (C 0)
				  (IP-SUB-SPE (NP-ACC *T*-1)
					      (NP-NOM (PRO^N +du))
					      (MDP wylle))))))


( (CODE )
  (IP-MAT (CONJ &)
          (CP-FRL-LFD (WNP-NOM-1 (WPRO^N locehwa))  <--- LOCAHW-
                      (C 0)
                      (IP-SUB (NP-NOM *T*-1)
                              (RP ut)
                              (VBPS gange)))
          (, ,)
          (VBPS licge)
          (NP-NOM-RSP (PRO^N he))
          (PTP-NOM (RP+VBN^N ofslagen))
          (. ,)) (ID Heptateuch,Josh:2.19.4381))

( (CODE )
  (IP-MAT-SPE (ADVP-TMP (ADV^T Nu))
              (NP-NOM (PRO^N ic))
              (MDP wylle)
              (NP-ACC-RFL (PRO^A me) 
                          (ADJP-ACC (ADJ^A sylfe)))
              (NP-DAT (PRO^D him))
              (ADVP (ADV so+dlice))
              (VB geoffrian)
              (, ,)
              (PP (P for+dan)
                  (CP-ADV-SPE (C 0)
			      (IP-SUB-SPE (NP-NOM (PRO^N ic))
					  (ADVP-TMP (ADV^T leng))
					  (NEG+HVP n+abbe)
					  (CP-FRL-SPE (WNP-ACC-1 (WPRO^A hw+at))  <--- WH-
						      (C 0)
						      (IP-SUB-SPE (NP-ACC *T*-1)
								  (NP-NOM (PRO^N ic))
								  (PP (P on)
								      (NP-DAT (PRO$ his) (N^D lacum)))
								  (VBD aspende))))))
              (. .)) (ID AelfLives,AELS[Lucy]:66.979))

( (IP-MAT (CONJ and)
          (NP-NOM *con*)
          (VBDI cw+a+d)
          (, :)
          (IP-MAT-SPE (NP-NOM (PRO^N Ic))
                      (VBP sece)
                      (NP-ACC (PRO$^A mine) (N^A gebro+dru))
                      (, ,)
                      (CP-FRL-LOC-SPE (WADVP-LOC-1 (WADV^L hwar))  <--- WH-
				      (C 0)
				      (IP-SUB-SPE (ADVP-LOC *T*-1)
						  (NP-NOM (PRO^N hig))
						  (VBPS healdon)
						  (NP-ACC (PRO$ heora) (N^A heorda)))))
          (. .)) (ID Heptateuch,Gen:37.16.1442))

Free relatives headed by +TE

There are a small number of relatives headed by +TE alone, which appear to be free relatives.


( (CODE )
  (IP-MAT (ADVP (ADV Yfele))
          (VBPI de+d)
          (NP-DAT-RFL-ADT (PRO^D him)
                          (ADJP-DAT (ADJ^D sylfum)))
          (CP-FRL-SBJ (WNP-NOM-1 0)
                      (C +te)                             <--- +TE
                      (IP-SUB (NP-NOM *T*-1)
                              (PP (P mid)
                                  (NP-DAT (N^D swicdome)))
                              (NP-GEN (PRO^G his))
                              (VBPI tila+d))) 
          (. ,)) (ID AelfLives,+ALS_[Alban]:172.2245))

(NODE (IP-MAT-SPE (ADVP-TMP (ADV^T Nu))
                  (BEPI synd)
                  (RP+VBN^N for+dfarene)
                  (CP-FRL-SBJ-SPE (WNP-NOM-1 0)
                                  (C +te)
                                  (IP-SUB-SPE (NP-NOM *T*-1)
                                              (NP (NP-GEN (D^G +d+as) (N^G cildes))
                                                  (N sawle))
                                              (VBDI sohton)))
                  (. .))
      (ID WSGospels,Mt_[WSCp]:2.21.112)) 

( (IP-MAT-SPE (CONJ And)
              (NP-NOM (D^N +ta) (N^N men))
              (NEG+VBPI nyton)
              (CP-FRL-SPE (WNP-NOM-1 0)
                          (C +te)
                          (IP-SUB-SPE (NP-NOM *T*-1)
                                      (PP (NP-DAT (PRO^D him))
                                          (P onufan))
                                      (VBPI ga+d)))
              (. .))
  (ID WSGospels,Lk_[WSCp]:11.44.4586))

In addition we do the following type as free relatives (with a reading something like you are the one who...). Note this only applies to cases with pronoun or demonstrative subjects. Similar cases with full NP subjects are presentational (there was a man who...) and are parsed as relatives (see Presentational there).


Context: Ne synd ge +te +t+ar spreca+d, ac spryc+t se halga gast on eow.

(NODE (IP-MAT-SPE (NEG Ne)
                  (BEPI synd)
                  (NP-NOM (PRO^N ge))
                  (CP-FRL-PRD (WNP-NOM-1 0)
                              (C +te)
                              (IP-SUB-SPE (NP-NOM *T*-1)
                                          (ADVP-LOC (ADV^L +t+ar))
                                          (VBPI spreca+d))))
      (ID AelfLives,+ALS[Lucy]:73.997)) 

(NODE (IP-MAT-SPE (NP-NOM (PRO^N Ic))
                  (BEPI eom)
                  (CP-FRL-PRD-SPE (WNP-NOM-1 0)
				  (C +te)
				  (IP-SUB-SPE (NP-NOM *T*-1)
					      (ADVP-LOC (ADV^L +d+ar))
					      (VBD syngode))))
      (ID AelfLives,+ALS[Pr_Moses]:257.1780)) 

( (IP-MAT-SPE (NP-NOM (D^N se))
              (BEPI is)
              (CP-FRL-PRD-SPE (WNP-NOM-1 0)
			      (C +de)
			      (IP-SUB-SPE (NP-NOM *T*-1)
					  (VBPI (VBPI sec+d) (, .) (CONJ and) (RP+VBPI tosc+at))))
              (. ;))
  (ID CathHomII,+ACHom_II,_13:131.122.2934))

( (IP-MAT-SPE (CONJ &)
              (NP-NOM (PRO^N hig))
              (BEPI synt)
              (CP-FRL-PRD-SPE (WNP-NOM-1 0)
			      (C +te)
			      (IP-SUB-SPE (NP-NOM *T*-1)
					  (NP (N gewitnesse))
					  (VBPI cy+ta+t)
					  (PP (P be)
					      (NP (PRO me)))))
              (. FULL-STOP))
  (ID WSGospels,Jn_[WSCp]:5.39.6102))

Reduced relatives (RRC)

PPCME2 Table of contents
YCOE Table of contents

*difference*

Most constituents labelled RRC by the PPCME2 are labelled as participle phrases (PTP) in the YCOE. Only reduced relatives not headed by participles are labelled RRC in the YCOE. These are almost all of the type X by (another) name.


(NODE (IP-SUB (NP-NOM (NPR^N Gaius)
                      (NP-NOM-PRN (D^N se) (N^N casere))
                      (, ,)
                      (RRC (NP-INS-ADT (ADJ^I o+dre) (N^I naman))
                           (NP-NOM (NPR^N Iulius))))
              (, ,)
              (NP-ACC (PRO^A hit))
              (PP (P mid)
                  (NP (N ferde)))
              (VBD (VBD gesohte) (CONJ &) (VBD geeode))
              (PP (NP-DAT-EXT (NUM^D syxtygum)
                              (NP-GEN (N^G wintra)))
                  (P +ar)
                  (NP (NP-GEN (NPR^G Cristes))
                      (N cyme))))
      (ID Bede1,Bede_1:2.30.13.61)) 

(NODE (IP-MAT (PP (P On)
                  (NP-DAT (NP-GEN (NPR^G Herodes)
                                  (NP-GEN-PRN *ICH*-1))
                          (N^D dagum)
                          (, ,)
                          (NP-GEN-PRN-1 (NP-GEN (NPR^G Iudea))
                                        (N^G cyninges))))
              (, ,)
              (BEDI w+as)
              (NP-NOM (Q^N sum) (N^N sacerd)
                      (RRC (PP (P on)
                               (NP (N naman)))
                           (NP-NOM-PRN (NPR^N Zacharias))))
              (PP (P of)
                  (NP-DAT (NPR Abian) (N^D tune)))
              (. ,))
      (ID WSGospels,Lk_[WSCp]:1.5.3539)) 

(NODE (IP-MAT (ADVP-LOC (ADV^L +D+ar))
              (BEDI w+as)
              (NP-NOM (Q^N sum) (N^N man)
                      (RRC (NP-TMP (NUMP (NUM eahta) (CONJ &) (NUM +trittig))
                                   (NP-GEN (N^G wintra)))
                           (PP (P on)
                               (NP (PRO$ his) (N untrumnysse)))))
              (. .))
      (ID WSGospels,Jn_[WSCp]:5.5.6038))

Clefts (CP-CLF)

There are only a few clear it-clefts in Old English, all of which have overt HIT. We also parse as clefts temporals of the type "(HIT) W+AS time-expression +T+AT..." found only in Bede (so far). For the the use of WADVP-TMP as the wh-operator, see Temporal and locative relatives.


(NODE (IP-SUB (NP-NOM (PRO^N hit))
              (BEDS w+are)
              (NP-NOM-PRD (D^N se) (N^N H+alend))
              (CP-CLF (WNP-NOM-1 0)
                      (C +te)
                      (IP-SUB (NP-NOM *T*-1)
                              (NP-ACC (PRO^A hyne))
                              (VBD h+alde))))
      (ID WSGospels,Jn_[WSCp]:5.15.6060)) 

(NODE (IP-SUB (NP-NOM (PRO^N hit))
              (BEDS w+are)
              (NP-NOM-PRD (NPR^N Swy+dun))
              (CP-CLF (CP-CLF (WNP-NOM-1 (D^N se))
                              (C +de)
                              (IP-SUB (NP-NOM *T*-1)
                                      (NP-ACC (PRO^A hine))
                                      (VBD l+arde)
                                      (PP (P mid)
                                          (NP-DAT (D^D +t+are) (ADJ^D halgan) (N^D lare)))))
                      (CONJP (CONJ and)
                             (CP-CLF (WNP-ACC-2 (D^A +tone))
                                     (C +de)
                                     (IP-SUB (NP-ACC *T*-2)
                                             (NP-NOM (PRO^N he))
                                             (VBDI geseah)
                                             (PP (P on)
                                                 (NP-DAT (D^D +d+are) (N^D cyrcan)))
                                             (ADJP-ACC (ADV swa) (ADJ^A f+agerne)))))))
      (ID AelfLives,+ALS_[Swithun]:388.2640)) 

(NODE (IP-MAT (NP-NOM *exp*)          <--- temporal cleft
              (ADVP-TMP (ADV^T +Da))
              (BEDI w+as)
              (PP (P ymb)
                  (NP (NUMP (NUM feower) (NUM hund)
                            (CONJP *ICH*-1))
                      (NP-GEN (N^G wintra))
                      (CONJP-1 (CONJ &) (NUM nigon) (CONJ &) (NUM feowertig))
                      (PP (P fram)
                          (NP (NP-GEN (PRO$^G ures) (NPR^G Drihtnes))
                              (N menniscnysse)))))
              (, ,)
              (CP-CLF (WADVP-TMP-2 0)
                      (C +t+at)
                      (IP-SUB (ADVP-TMP *T*-2)
                              (NP-NOM (NPR^N Martianus) (N^N casere))
                              (NP (N rice))
                              (RP+VBDI onfeng))))
      (ID Bede1,Bede_1:12.50.17.283))

Infinitival clauses (IP-INF, IP-INF-NCO)

Non-complement infinitives
Infinitival subjects in situ
Accusative and infinitive constructions (ECM verbs)
Passives with infinitives (THE BOOK IS TO READ)
Tough-movement and similar constructions
Infinitival relatives

PPCME2

Non-complement infinitives

*difference*
Any infinitive which is not a complement, usually but not exclusively purpose infinitives, is labelled IP-INF-NCO.


( (IP-MAT-SPE (CONJ &)
              (NP-NOM (PRO^N we))
              (VBDI comon)
              (NP-RFL-ADT (PRO us))
              (IP-INF-NCO-SPE (NP-DAT (PRO^D him))
			      (TO to)
			      (VB^D geeadmedenne))
              (. .)) (ID GospMt,Mt_[WSCp]:2.2.67))

( (IP-MAT-SPE (CONJ &)
              (VBI binda+t)
              (NP-DAT-ADT (N^D sceafm+alum))
              (IP-INF-NCO-SPE (TO to)
			      (VB^D forb+arnenne))
              (. FULL-STOP)) (ID GospMt,Mt_[WSCp]:13.30.846))

( (IP-MAT-SPE (VBP Sylle)
              (NP-NOM (PRO^N ge))
              (NP-DAT (PRO^D him))
              (IP-INF-NCO-SPE (VB etan))
              (. .)) (ID GospMt,Mt_[WSCp]:14.16.935))

Infinitival subjects in situ

Like other clausal subjects in situ (see Clausal subjects), infinitival subjects are given the extended label -SBJ.


( (IP-MAT-SPE (IP-INF-SBJ-SPE-3 (TO To)
				(VB^D sittanne)
				(PP (PP (P on)
					(NP-ACC (PRO$^A mine) (ADJ^A swi+tran) (N^A healfe)))
				    (CONJP (CONJ o+d+de)
					   (PP (P on)
					       (NP-ACC (ADJ^A wynstran))))))
              (NEG+BEPI nys)
              (NP-DAT (NP-DAT (PRO^D me))
                      (CONJP *ICH*-1))
              (IP-INF-SPE (IP-INF-SPE *-3)
			  (NP (PRO inc))
			  (TO to)
			  (VB^D syllanne))
              (CONJP-1 (CONJ ac)
                       (NP-DAT (D^D +tam)
                               (CP-REL-SPE (WNP-2 0)
					   (C +te)
					   (IP-SUB-SPE (NP *T*-2)
						       (NP-NOM (PRO^N hyt))
						       (PP (P fram)
							   (NP-DAT (PRO$^D minum) (N^D F+ader)))
						       (VBN gegearwod)
						       (BEPI ys)))))
              (. .)) (ID GospMt,Mt_[WSCp]:20.23.1333))

Accusative and infinitive constructions (ECM verbs)

Verbs on the following list (from Calloway) take the accusative and infinitive construction (ECM verbs). Arbitrary PRO subjects are not indicated (unlike in the PPCME2) but overt subjects have the -SBJ extended label to distinguish them from other arguments.

behealdan, (a)biddan, (a)findan, hatan, (a/for)l+atan, (ge)don, geascian, gefelan, gefrignan, gehawian, gemetan, gemittan, ge+dafian, geunnan, gewitan, (ge)hieran, l+aran, ongietan, sceawian, (of/ge)seon, talian, tellan


( (IP-MAT (CONJ and)
          (NP-NOM *con*)
          (VBDI het)
          (IP-INF (NP-ACC-SBJ (PRO^A hine))
                  (ADVP-TMP (ADV^T symble))
                  (BE beon)
                  (PP (P +atforan)
                      (NP (PRO$ his) (N gesih+de))))
          (. .)) (ID AelfLives,+ALS_[Sebastian]:13.11))

( (CODE )
  (IP-MAT-SPE (VBI L+ata+d)
              (NP-ACC-1 (PRO^A hi))
              (ADVP-TMP (ADV^T nu))
              (IP-INF-SPE (NP-ACC-SBJ *ICH*-1)
			  (VB faran)
			  (PP (P to)
			      (NP-DAT (D^D +dam) (ADJ^D forestihtan) (N^D kynehelme))))
              (. ,)) (ID AelfLives,+ALS_[Sebastian]:80.49))

( (IP-MAT (CONJ and)
          (NP-NOM *con*)
          (VBDI forleton)
          (IP-INF (NP-ACC-SBJ (PRO^A hine))
                  (ADVP (ADV swa))
                  (VB licgan)
                  (PP (P for)
                      (NP-ACC (ADJ^A deadne))))
          (. .)) (ID AelfLives,+ALS_[Sebastian]:425.258))

( (CODE )
  (IP-MAT (ADVP-TMP (ADV^T +Ta))
          (PP (P on)
              (NP-DAT (Q^D sumere) (N^D nihte)))
          (VBDI gesawon)
          (NP-NOM (PRO^N hi))
          (IP-INF (VB cuman)
                  (NP-ACC-SBJ (NP-ACC (Q^A mycel) (ADJ^A m+adenlic) (N^A werod))
                              (, ,)
                              (CONJP (CONJ and)
                                     (NP-ACC (NPR^A Agnes)
                                             (ADVP-LOC (ADV^L tomiddes))))))
          (. .)) (ID AelfLives,AELS[Agnes]:250.665))

Passives with infinitives (THE BOOK IS TO READ)

In clauses with an active infinitive with passive meaning where the subject of the main clause is also the object of the infinitive, the subject is traced with an A-movement trace to the infiniitve. The trace in the infinitive is not marked for case.


(NODE (IP-SUB-SPE (NP-NOM-1 (D^N se) (ADJ^N trywleasa))
                  (NEG ne)
                  (BEPI bi+d)
                  (NP-DAT (NEG+Q^D nanum) (N^D hlaforde))
                  (IP-INF-SPE (NP *-1)
                              (TO to)
                              (HV^D h+abbenne)))
      (ID AelfLives,+ALS[Ash_Wed]:129.1550)) 

(NODE (IP-MAT (ADVP (ADV For+don))
              (NEG ne)
              (BEPI seondon)
              (IP-INF (TO to)
                      (VB^D lufienne)
                      (NP *-1))
              (NP-NOM-1 (D^N +ta) (N^N wiisan))
              (PP (P fore)
                  (NP-DAT (N^D stowum)))
              (. ,))
      (ID Bede1,Bede_1:16.66.24.447))

See also the IS TO WIT THAT construction.

Tough-movement and similar constructions

Tough-movement constructions (John is easy to please) are done in a similar way to other passives with infinitives. The subject is traced with an A-movement trace to the infinitive. The infinitive is contained within the ADJP. Similar but perhaps not identical constructions, like The city is fair to see, are done in the same way.


(IP-SUB (NP-NOM-1 (D^N +d+at) (N^N treow))
	(BEDI w+as)
	(ADJP-NOM-PRD (ADJ^N god)
		      (IP-INF (NP *-1)
			      (TO to)
			      (VB^D etenne))))


(IP-MAT (NP-NOM-1 (D^N +T+at))
	(BEPI is)
	(ADJP-NOM-PRD (ADJ^N ungeliefedlic)
		      (IP-INF (NP *-1)
			      (TO to)
			      (VB^D gesecganne))))

In the alternative construction (it is easy to please John) the expletive HIT is coindexed to the infinitive in the usual way (see Expletive constructions).

Infinitival relatives

Infinitival relatives are treated in essentially the same way as in the PPCME2. Note that infinitives inside CP-EOP are labelled as non-complement infintives (-NCO). When separated from its antecedent it is traced, as usual.


( (CODE )
  (IP-MAT (CONJ &)
          (NP-NOM (PRO^N hig))
          (VBDI sealdon)
          (NP-DAT (PRO^D hym))
          (NP-ACC (N^A win)
                  (CP-EOP (WNP-1 0)
                          (IP-INF-NCO (NP *T*-1)
                                      (VB drincan)))
                  (PTP-ACC (PP (P wi+d)
                               (NP (N eallan)))
                           (VBN^A gemenged)))
          (. FULL-STOP)) (ID WSGospels,Mt_[WSCp]:27.34.2044))

(CP-QUE-SPE (WADVP-3 (WADV hu))
	    (IP-SUB-SPE (ADVP *T*-3)
			(MDPI m+ag)
			(NP-NOM (D^N +tes)) 
			(NP-ACC (PRO$ his) (N^A fl+asc)
				(CP-EOP-SPE *ICH*-1))
			(NP (PRO us))
			(VB syllan)
			(CP-EOP-SPE-1 (WNP-2 0)
				      (IP-INF-NCO-SPE (NP *T*-2)
						      (TO to)
						      (VB^D etene)))))

Small clauses (IP-SMC)

Small clauses with passivized subject
Tracing to small clauses
Small clauses with present participles
PPCME2 Table of contents
YCOE Table of contents

PPCME2

The predicate of a small clause if not verbal is always indicated by the -PRD label; the subject is either nominative (-NOM) or indicated with -SBJ if non-nominative or ambiguous for case.


( (IP-MAT (CONJ and)
          (NP-NOM *con*)
          (VBDI gemetton)
          (IP-SMC (NP-ACC-SBJ (PRO^A hine))
                  (ADJP-ACC-PRD (ADJ^A untrumne)))
          (. ,)) (ID AelfLives,+ALS_[Sebastian]:256.155))

( (IP-MAT (CONJ and)
          (NP-NOM *con*)
          (VBD gemette)
          (IP-SMC (NP-ACC-SBJ (PRO^A hine))
                  (VAG^A libbendne))
          (. .)) (ID AelfLives,+ALS_[Sebastian]:430.263))

( (IP-MAT (CONJ and)
          (NP-NOM *con*)
          (VBD ar+arde)
          (NP-ACC (D^A +t+at) (N^A mynster))
          (QP (Q eall))
          (PP (P be)
              (NP-DAT (NP-GEN (NPR^G Maures))
                      (N^D r+ade)))
          (PP (P wi+d)
              (NP-ACC (D^A +ta) (Q^A mycclan) (N^A ea)
                      (CP-REL (WNP-1 0)
                              (C +te)
                              (IP-SUB (NP-NOM (N^N menn))
                                      (VBPI hata+d)
                                      (IP-SMC (NP-SBJ *T*-1)   <--- ambiguous
                                              (NP-PRD (NPR Liger)))))))
          (. .)) (ID AelfLives,+ALS_[Maur]:157.372))

( (IP-MAT (CONJ and)
          (ADVP-LOC (ADV^L +d+ar))
          (NEG+BEDI n+as)
          (NEG+ADV na)
          (NP-NOM (Q ma)
                  (CP-REL (WNP-NOM-1 0)
                          (C +te)
                          (IP-SUB (NP-NOM-2 *T*-1)
                                  (IP-SMC (NP-NOM *-2)          <--- nominative
                                          (ADJP-NOM-PRD (ADJ^N manful)))
                                  (NEG+BEDS n+are)
                                  (VBN gemet))))
          (. .)) (ID AelfLives,+ALS[Pr_Moses]:207.1755))

Small clauses with passivized subject

When the subject of a small clause has been passivized to the subject position of the matrix clause, the case of the predicate in the small clause is always nominative and is so marked even when it is morphologically ambiguous. Passivization is indicated in this case by an A-movement trace since an argument has moved across a clause boundary.


(NODE (IP-SUB (NP-NOM-2 (PRO^N he))
              (PP (P +turh)
                  (NP-ACC (PRO$ his) (N^A bebod)))
              (IP-SMC (NP-NOM *-2)
                      (ADJP-NOM-PRD (ADJ^N cristene)))
              (VBN gefremed)
              (BEDS w+are))
      (ID Bede1,Bede_1:4.32.5.78)) 

(NODE (IP-SUB (NP-NOM-2 *T*-1)
              (ADVP-LOC (ADV^L +d+ar))
              (IP-SMC (NP-NOM *-2)
                      (ADJP-ACC-PRD (ADJ^A regollico)))
              (VBN gedemed)
              (BEDI w+aron))
      (ID Bede2,Bede_2:4.108.13.157))

Note that certain verbs (GEWEOR+TAN, +TENCAN, GEHATAN, etc.) are treated as taking a predicate directly in the passive, rather than taking a small clause. See Passive verbs taking predicates.


(IP-MAT (BEDI W+as)
	(NP-NOM (D^N se) (ADJ^N grimma) (N^N g+ast))
	(NP-NOM-PRD (NPR^N Grendel))
	(VBN haten)
	(, ,))

Cf. constructions with HATTE where both arguments are nominative.

Tracing to small clauses

Tracing to small clauses follows the policy of the Brooklyn Corpus:

If the two constituents of the small clause are adjacent and the subject is before the predicate, they are bracketed as they appear within small-clause brackets, no matter where in the matrix clause the small clause appears: (IP-SMC (SUBJECT) (PREDICATE)) ( (IP-MAT (CONJ and) (NP-NOM *con*) (VBDI gemetton) (IP-SMC (NP-ACC-SBJ (PRO^A hine)) (ADJP-ACC-PRD (ADJ^A untrumne))) (. ,)) (ID AelfLives,+ALS_[Sebastian]:256.154)) If the two constituents of the small clause are separated, and the subject is before the predicate, the small-clause bracketing appears with the predicate and the subject is traced: (IP (SUBJECT-1) ... (IP-SMC (SUBJECT *ICH*-1) (PREDICATE))) (NODE (IP-SUB (NP-NOM (PRO^N he)) (NP-ACC-1 (Q^A sum) (N^A +ting) (CP-REL *ICH*-2)) (HVD h+afde) (IP-SMC (NP-ACC-SBJ *ICH*-1) (ADJP-ACC-PRD (ADJ^A untobrocen))) (, ,) (CP-REL-2 (WNP-NOM-3 0) (C +te) (IP-SUB (NP-NOM *T*-3) (NP (PRO$ his) (N h+ale)) (VBD hremde) (PP (P +turh) (NP-ACC (ADJ^A re+de) (N^A wiglunga)))))) (ID AelfLives,+ALS_[Sebastian]:256.155)) If the predicate of the small clause appears before the subject and the subject appears after the matrix verb, the small clause contains the subject and therefore also appears after the matrix verb: (IP (PREDICATE-1) ... (VB) (IP-SMC (PREDICATE *ICH*-1) (SUBJECT))) ( (IP-MAT (CONJ and) (NP-NOM *con*) (NP-DAT-ADT (PRO^D heom)) (NP-ACC-PRD-1 (N^A weardas)) (VBDI setton) (IP-SMC (NP-ACC-PRD *ICH*-1) (NP-ACC-SBJ (ADJ^A +turhwacole) (N^A menn))) (...) (ID AelfLives,+ALS[Forty_Soldiers]:145.1324))) If the predicate of the small clause appears before the subject and the subject precedes the matrix verb, the small clause contains two traces and appears directly following the matrix verb: (IP (PREDICATE-1) (SUBJECT-2) ... (VB) (IP-SMC (SUBJECT *ICH*-2) (PREDICATE *ICH*-1)))

Small clauses with present participles

Note that clauses of the type Mary saw Sue running down the street are technically ambiguous since either Mary or Sue could be the subject of the predicate running down the street. When Mary is the subject, running down the street is a participle phrase; but when Sue is the subject then Sue running down the street is a small clause.

*difference
In the PPCME2 this ambiguity was always resolved in favour of the adjunct participial structure. In the YCOE, clauses are assigned the different structures based on context; unclear cases are labelled as adjunct participials.


(NODE (IP-MAT (CONJ &)
              (PP (P +ta)
                  (CP-ADV (C 0)
                          (IP-SUB (NP-NOM (PRO^N he))
                                  (NP-ACC-RFL (PRO^A hine))
                                  (ADVP-TMP (ADV^T eft))
                                  (RP agen)
                                  (VBD bewende))))
              (, ,)
              (NP-NOM (PRO^N he))
              (VBDI funde)
              (IP-SMC (NP-ACC-SBJ (PRO^A hi))
                      (VAG sl+apende))
              (. ;))
      (ID WSGospels,Mk_[WSCp]:14.40.3312))

Absolutes (PTP-case-ABS)

Infinitive absolutes
PPCME2 Table of contents
YCOE Table of contents

Infinitive absolutes

So far, there is only one example of an infinitival absolute.


( (CODE )
  (IP-MAT (NP-NOM (PRO^N Hie))
          (NP (PRO$ heora) (N here))
          (PP (P on)
              (NP-ACC (NUM^A tu)))
          (RP+VBDI tod+aldon)
          (, ;)
          (IP-INF-ABS (IP-INF-ABS (NP-NOM (ADJ^N o+ter)) 
                                  (PP (P +at)
                                      (NP-ACC (N^A ham)))
                                  (BE beon)
                                  (IP-INF-NCO (NP-ACC (PRO$ heora) (N^A lond))
                                              (TO to)
                                              (VB^D healdanne)))
                      (, ,)
                      (CONJP (IP-INF-ABS (NP-NOM (ADJ^N o+der))
                                         (RP ut)
                                         (VB faran) 
                                         (IP-INF-NCO (TO to)
                                                     (VB^D winnanne)))))
          (. .)) (ID Orosius,:10.30.2.482))