Pairing Word Order With Headedness in the History of English

Goal • Goal: check the distributional tendencies affecting the placement of modifiers/adjuncts and complements/arguments in noun phrases, adjectives phrases and verb phrases Pairing word order with headedness in the history of English: a corpus-based analysis • Examples: (1) Now I will deal [with the construction] [in a way which will lead to odd results]. (1’) Now I will deal [in a way which will lead to odd results] [with the construction]. (2) the author [of this book] [from London] [N.A. the author is from London] (2’) the author [from London] [of this book] (3) keen [on music] [to a large extent] (3’) keen [to a large extent] [on music] 3 Background Outline • Complements – semantically selected or subcategorised • goal • background • description of the corpora: complementation and adjunction strategies • analysis of the data: – Complements-first – End-weight • Matthews (2007: 187): “unit in a construction either required or specifically taken by an individual member of a lexical category” • Matthews (1981: 124-127): impossibility of dropping (if dropped, then latent) • exclusion when the pattern is saturated – syntactic dependencies: • lexical restrictions or formal determination (Greenbaum et al. 1996: 76) {deal, compliance} + with-PP {assume, certain, hypothesis} + that-clause • concluding remarks • Adjuncts – loose semantic connection between the adjunct and the head => not required 2 4 Background The corpora • Distribution of complements and adjuncts: • Old English: 1.5+ million words (Old English section of the Diachronic Part of the Helsinki Corpus of English Texts, with certain additions, c750–) – syntactic explanation: • Quirk et al. (1985: 49-50): ‘Complements first’ • Hawkins (2007): ‘Arguments precede X’ – processing explanation: • Quirk et al. (1985: 1398): End-weight • Hawkins’ (2004) ‘Minimize Domains’ or MiD: Given two or more categories A, B, [...] related by a grammatical rule R of combination and/or dependency, the human processor prefers to minimize the distance between them within the smallest surface structure domain sufficient for the processing of R. (Hawkins 2000: 234) Hawkins (2007) hypothesises that MiD is relevant especially to examples of complementation. Taylor, Ann, Anthony Warner, Susan Pintzuk and Frank Beths (2003) The York-Toronto-Helsinki Parsed Corpus of Old English Prose. • Middle English: 1,155,965 words (Middle English section of the Diachronic Part of the Helsinki Corpus of English Texts, with certain additions and deletions, 1150–1500) Kroch, Anthony and Ann Taylor (2000) Penn-Helsinki Parsed Corpus of Middle English, second edition. • Early Modern English: 1,737,853 words (the Helsinki directories of the Penn-Helsinki Parsed Corpus of Early Modern English plus two supplements; 1500–1710) Kroch, Anthony, Beatrice Santorini and Lauren Delfs (2004) Penn-Helsinki Parsed Corpus of Early Modern English. • (Late) Modern English: 948,895 words (1700–1914) 5 Background Kroch, Anthony, Beatrice Santorini and Ariel Diertani (2010) Penn Parsed Corpus of Modern British English. 7 The corpora • Examples: (1) Now I will deal [with the construction] [in a way which will lead to odd results]. (1’) Now I will deal [in a way which will lead to odd results] [with the construction]. (1) is claimed to be a better performance solution, on processing grounds (MiD, end-weight), than (1’) because of the amount of structure which has to be processed (between the head category and the second constituent in the (local) phrase). 6 • periods: OE, ME, EModE and (L)ModE • parsed corpora, with (almost) identical similar parsing conventions • parsed files (.psd), using P&P-based part-of-speech and syntactic tags • retrieval by means of CorpusSearch • (extensive) manual revision ./.. 8 but, if you approve of this, if you please to lett me know y=r= pleasure, I will tell it M=r= Isaac. (ANHATTON-E3-H,2,214.41) ((IP-MAT (CONJ but) (, ,) (PP (P if) (CP-ADV (C 0) (IP-SUB (, ,) (PP (P if) (CP-ADV (C 0) (IP-SUB The data QP + OBJ neither will I againe smite any more euery thing liuing, as I haue done. (AUTHOLD-E2-H,VIII,20G.466) (NP-SBJ (PRO you)) (VBP approve) (PP (P of) (NP (D this)))))) PP + OBJ and sitting in some place, where no man shall prompe him, by him self, let him translate into Englishe his former lesson. (ASCHE1-H,1V.22) ADJ + OBJ Lisle. My Lord, this Fellow that now speaks against me, broke open my Trunk, (LISLE-E3-H,IV,120C1.203) (NP-SBJ (PRO you)) (VBP please) (IP-INF (TO to) (VB lett) (IP-INF (NP-SBJ (PRO me)) (VB know) (NP-OB1 (PRO$ y=r=) (Npleasure))))))) (, ,) (NP-SBJ (PRO I)) (MD will) (VB tell) (NP-OB1 (PRO it)) (NP-OB2 (NPR M=r=) (NPR Isaac)) (. .)) ADV + OBJ Moreouer, there is no one thing, that hath more, either dulled the wittes, or taken awaye the will of children from learning, then the care they haue, to satisfie their masters, in making of latines. (ASCH-E1-H,1R.9) 9 The data Examples of queries • Verb-governed contexts: V + adjunct + complement verb group immediately precedes an adjunct, and the adjunct immediately precedes an object OE: node: IP* query: (VB* iprecedes W*|NUMP*|QP*|PP*|ADJ*|ADV*|INTJP|XP| NP-ADT*|NP-DIR*|NP-EXT*|NP-LOC*|NP-TMP*| NP-VOC*) AND (W*|NUMP*|QP*|PP*|ADJ*|ADV*|INTJP|XP| NP-ADT*|NP-DIR*|NP-EXT*|NP-LOC*|NP-TMP*| NP-VOC* iprecedes NP-ACC*|NP-DAT*) ME, EModE and ModE: node: IP* query: (VB* iprecedes W*|QP|PP|RRC|ADJ*|ADV*| CP-*|IP-SUB|NP-*) AND (W*|QP|PP|RRC|ADJ*|ADV*|CP-*|IP-SUB| NP-* iprecedes NP-OB*) 10 NP-ADV + OBJ: sends this way one, That way another, (MIDDLET-E2-H,21.513) 11 The data • Verb-governed contexts: V + complement + adjunct verb group immediately precedes an object, and the object immediately precedes an adjunct OE: node: IP* query: (VB* iprecedes NP-ACC*|NP-DAT*) AND (NP-ACC*|NP-DAT* iprecedes W*|NUMP*|QP*|PP*|ADJ*|ADV*|INTJP|XP| NP-ADT*|NP-DIR*|NP-EXT*|NP-LOC*| NP-TMP*|NP-VOC*) ME, EModE and ModE: node: IP* query: (VB* iprecedes NP-OB*) AND (NP-OB* iprecedes W*|QP|PP|RRC|ADJ*|ADV*|CP-*|IP-SUB|NP-*) 12 The data The data • Adjective-/Noun-governed contexts: A/N + adjunct + complement head (adjective/noun) immediately precedes an adjunct, and the adjunct immediately precedes a complement (that- or infinitive clause) OBJ + QP so this time will trouble y=r= Losp no more w=th= y=r= most obedient, duttyful daughter, A. Nottingham. (ANHATTON-E3H,2,212.29) OBJ + PP I thoughte I wolde take some spendyng money wyth me (MERRYTAL-E1-H,31.148) OBJ + RRC If you have provided those Stockings before spoken of I pray send them. (STRYPE-E3-H,183.69) OBJ + ADJ and cut it not so close to the Body as to hurt it, nor yet so long that it be a Stump, (LANGF-E3-H,122.269) EModE and ModE: node: NP* query: (NP* iDoms CP-THT) AND (NP* iDoms *P*) AND (*P* iprecedes CP-THT) node: NP* query: (NP* iDoms IP-INF) AND (NP* iDoms *P*) AND (*P* iprecedes IP-INF) 13 The data 15 The data OBJ + ADV But my Brother understood the matter aright (HOXINDEN-1660E3-H,280.162) OBJ + IP-SUB Contrariwise, there be others againe, that will not endure to read a short composition, bee it neuer so well approoued: (CLOWES-E2H,34.315) OBJ + NP-ADV I 'll ply him that way, (FARQUHAR-E3-H,9.326) 14 PP + CP-THT/IP-INF the duke had got a solemn promise of the king that he would never speak to him of religion. (BURNETCHA-E3-P2,2,180.98) The master shewyng us that by neglygence of some to belay the haylers, the mayn yerde had fawln down (MADOX-E2P1,112.434) hee is able without great care, study and much diligence, to labour commendably, (CLOWES-E2-P1,8.69) RC + CP-THT/IP-INF [He would not hearken to this, which made me inclined to believe] a report I had heard, that the duke had got a solemn promise of the king that he would never speak to him of religion. (BURNETCHA-E3-P2,2,180.98) Ond sona fylgde +t+at weorc, +t+at him And immediately carried out the(neu.) work(neu.) rel(neu) him gewunelic w+as, +t+at he godspellice lare l+arde; accustomed was that he evangelical teaching taught in mind of the great Obligation that lies on them to live sutably to their Profession: (BURNETROC-E3-P2,122.170) 16 The data The data adjunct + CP-THT And there was a feeling by no means uncommon, and very deadly, that India would be lost for ever, and with it all the glory of England. (TROLLOPE-1882,177.356) yea I am sorie, with all my harte, that they be giuen no more to riding, then they be: (ASCH-E1-P1,10R.186) IP-INF + than/that(so...that) for I hadde levyr hafe bene in that payne to domysdaye for I had better have been in the pain to Doomsday than hafe comen to hevene (CMJULNOR,55.163) than have come to heaven euerie man is more readie to argue, than to obey and execute. (BACON-E2-P1,1,7R.73) but the Bassa here was in such haste to see us, that he dispatched the courier back which Mr. W- had sent to know the time he would send the convoy to meet us without suffering him to pull off his boots. (MONTAGU-1718,86.111) CP-THT + than For we are no less certain that there is a great Town called Constantinople, the seat of the Ottoman Empire, than that there is another called London. (BURNETROC-E3-P1,79.231) & beon geornran +t+at we Godes bebodu healdan, +tonne we & be more eager that we God’s command keep than we urne teonan gewrecan. (coblick,HomS_10_[BlHom_3]:33.123.444) our harm avenge 17 The data 19 The data • Adjective-/Noun-governed contexts: A/N + complement + adjunct head (adjective/noun) immediately precedes a complement (that- or infinitive clause), and the complement immediately precedes an adjunct EModE and ModE: node: NP* query: (NP* iDoms CP-THT) AND (NP* iDoms *P*) AND (CP-THT iprecedes *P*) CP-THT/IP-INF + sentential relative clause There is a wise saying that nine-tenths of the noble work done in the world is drudgery, which is often misused as if it meant that nine-tenths of the drudgery done in the world is noble work. (BENSON-1908,46.109) [King James sent a Person down to him, with] Offers to mitigate his Fine upon Conditions of ready Payment, to which his Lordship reply'd, that if his Majesty pleas'd to allow him a little longer time, he would rather chuse to play double or quit with him: (CIBBER1740,44.134) CP-THT + apposition he also expressed an opinion that mulattoes inherited the vices of both races - a maxim which I had heard often enough before, (READE-1863,225.616) node: NP* query: (NP* iDoms IP-INF) AND (NP* iDoms *P*) AND (IP-INF iprecedes *P*) 18 20 Analysis of the data Analysis of the data • Verb-governed contexts: 100% 80% (i) Syntactic principle ‘Complements-first’ 60% c ompl_last c ompl_f irst 40% 20% 0% OE •Statistical •Statistical •Statistical •Statistical significance significance significance significance ME for for for for EM odE M odE full variation: yes (P<.0001) variation OE>ME: no (P=0.0949) variation ME>EModE: yes (P<.0001) variation EModE>ModE: yes (P<.0001) 21 Analysis of the data 23 Analysis of the data • Verb-governed contexts: • Adjective-governed contexts: (based on Pérez-Guerra and Martínez-Insua 2010a,b) compl_first compl_last OE 8270 3290 ME 17803 6790 EModE 7859 1995 ModE 13084 2579 compl_first compl_last 22 OE 10 0 ME EModE 20 19 31 29 ModE 32 28 24 Analysis of the data Analysis of the data • Adjective-governed contexts: • Noun-governed contexts: 100% 100% 90% 90% 80% 80% 70% 70% 60% 60% compl_last 50% compl_first 40% 30% 30% 20% 20% 10% 10% 0% 0% OE •Statistical •Statistical •Statistical •Statistical compl_last 50% compl_first 40% significance significance significance significance ME for for for for EModE variation variation variation variation ModE OE OE>ME: n.a. ME>EModE: no (P=0.8625) EModE>ModE: no (P=0.2191) ME>ModE: no (P=0.2299) •Statistical •Statistical •Statistical •Statistical significance significance significance significance ME for for for for EModE variation variation variation variation ModE OE>ME: n.a. (apparently significant) ME>EModE: no (P=0.8415) EModE>ModE: no (P=0.1444) ME>ModE: no (P=0.2311) 25 27 Analysis of the data Analysis of the data • Noun-governed contexts: 100 compl_first compl_last OE 1 22 ME 20 86 EModE 26 105 ModE 15 107 90 80 70 60 ver b 50 adject ive noun 40 30 20 10 0 1 1: 2: 3: 4: 26 2 3 4 OE ME EModE ModE 28 Analysis of the data Analysis of the data 100 90 80 70 • VPs: 60 verb 50 adject ive noun 40 30 20 10 – most are complement-first – ME>ModE: (statistically significant) increase of complement-first VPs – pivotal period: ME>EModE 0 1 2 3 4 (ii) Processing principle ‘End-weight’ • APs: – OE: all the examples are complement-first (few data) – after ME: half are complement-first – pivotal period: ME • NPs: – most are complement-last (ME>ModE variation is not significant) – pivotal period: ME 29 31 Analysis of the data Analysis of the data Complement-first VPs: (no. of words, outliers excl.) So… • Heads and compliance with Complementsfirst: 70 60 50 40 30 20 10 0 – VPs clearly comply with Complements-first. – NPs do not comply with Complements-first. – APs occupy an intermediate position since half comply with Complements-first. – connection head status
Complements-first min max mean • Chronology of variation: – VPs: increase of complement-first from ME onwards – NPs: complement-last in all periods; quantitative stability from ME onwards – APs: intermediate situation from ME onwards 30 - End weight:
in all periods 32 Analysis of the data Analysis of the data Complement-last VPs: (no. of words, outliers excl.) 70 60 50 40 30 20 10 0 Complement-last APs: (no. of words, outliers excl.) 70 60 50 40 30 20 10 0 min max min max mean mean - End-weight: , less noticeable across time - End-weight not evident, more noticeable in ModE 33 35 Analysis of the data Analysis of the data Complement-first APs: (no. of words, outliers excl.) 70 60 50 40 30 20 10 0 Complement-first NPs: (no. of words, outliers excl.) 70 60 50 40 30 20 10 0 min max mean min max mean - End weight:  in all periods; more noticeable in ModE - End-weight is not clear in the periods investigated 34 36 Analysis of the data Analysis of the data Complement-last NPs: (no. of words, outliers excl.) 70 60 50 40 30 20 10 0 So… • general End-weight accomplishment hierarchy: VA>N: • frequency: the number of bare (intransitive) nouns in NPs outweighs the number of bare (intransitive) VPs • paradigmatic functionality: verbs take part in a wider range of complementation patterns than nouns • ellipsed nouns in NPs are more common than ellipsed verbs in VPs • morphological possibilities (Givón 1993: morphology implies syntactic integration) – APs: intermediate position since adjectives are more verbal than nouns 38 40 Concluding remarks Concluding remarks • Competing forces: syntax (Complements-first) and processing (End-weight): – connection complement distribution and head category: the more verbal the head is, the less likely its compliance with End-weight – when syntax fails, End-weight is crucial and justifies complement-last examples with the three categories (radically with nouns), especially in ModE – when syntax is at work (Complements-first), End-weight is not responsible for the distribution of complements and adjuncts since End-weight is not necessary 41 Concluding remarks • Competing forces: syntax (Complements-first) and processing (End-weight): The findings are more noticeable from EModE onwards. • word order: OE: “word order regulated by pragmatic tendencies [authors: path “towards one determined predominantly by syntactic rules”]” (Van Hoorick 1994: 53), with some trends: • V2 (several possibilities in the pre-V slot) • light-heavy distribution (Traugott 1992: 276) ME: fixation of the distribution of constituents in the clause in progress (Fischer 1992: 371). EModE: fixed (syntacticised) word order 42 • So... “the biggest single predictor of relative orderings (…) is syntactic weight” (Hawkins 2000: 232) is not strictly correct since, according to the data, syntax (Complements-first) seems to be a bigger predictor. [“in general the light-heavy distribution [end-weight] is no longer a major factor in English word order” (Traugott 1992: 276)] 43