The Northern Subject Rule in the present plural indicative in Older Scots. A corpus study of the Linguistic Atlas of Older Scots (1380-1500)

1 RODRIGUEZ LEDESMA Nombre de reconocimiento (DN): c=ES, MARIA NIEVES - 30532034V Fecha: 2024.10.09 22:29:26 +02'00' Estudios Lingüísticos, Literarios y Culturales (MELLC) The Northern Subject Rule in the present plural indicative in Older Scots. A corpus study of the Linguistic Atlas of Older Scots (1380-1500) ANEXO ÚNICO PORTADA Y DECLARACIONES del TFM MÁSTER UNIVERSITARIO EN TÍTULO DEL TRABAJO DE FIN DE MÁSTER AUTOR DEL TFM Nº DNI o pasaporte Correo-e Ana Ayala Díaz 77931074J [email protected] Con su firma, el estudiante SOLICITA, DECLARA y AUTORIZA lo siguiente: SOLICITUD DE DEPÓSITO (ANEXO IV) SOLICITO que sea aceptado el Trabajo de Fin de Máster arriba reseñado en calidad de DEPÓSITO, con objeto de poner en marcha los trámites de defensa en los plazos que correspondan. DECLARACIÓN DE ORIGINALIDAD (ANEXO I) (NRTFE: Art. 8.1) DECLARO 1. que el presente Trabajo de Fin de Máster es original e inédito y está libre de plagios; 2. que ha citado adecuadamente los textos y ejemplos tomados de otros autores; 3. que es consciente de las consecuencias académicas y/o disciplinarias derivadas del incumplimiento de las dos declaraciones anteriores, en aplicación de las normas vigentes de la Universidad de Sevilla. AUTORIZACIÓN DE CONSULTAS (ANEXO V) AUTORIZO/NO AUTORIZO que el Trabajo de Fin de Máster se incorpore al repositorio de la Universidad de Sevilla para que se ponga a disposición y consulta para potenciales interesados. Márquese una opción: AUTORIZO NO AUTORIZO Firmado: DECLARACIÓN DE CONFORMIDAD DEL TUTOR/A (ANEXO I) El profesor/a D./Dña. María Nieves Rodríguez Ledesma, del Departamento de Filología Inglesa (Lengua Inglesa), tutor/a del estudiante autor/a de este Trabajo de Fin de Máster, EXPRESA su conformidad con su contenido y con su forma, y autoriza los trámites administrativos para su depósito y defensa. Igualmente EXPRESA no ser consciente ni responsable de posibles plagios ni del uso indebido de textos o ideas de otros autores. Firmado: _ (el Tutor /a) FACULTAD DE FILOLOGÍA. Universidad de Sevilla. C/ Palos de la Frontera s/n 41004 FECHA: 09/10/2024 2 Abstract: This study focuses on how the Northern Subject Rule functions in the first, second, and third person plural contexts of the present indicative in Older Scots. It confirms that these plural forms were subject to the NSR between 1380 and 1500 based on the analysis of the corpus Linguistic Atlas of Older Scots, and it investigates the interrelated behaviour of the Proximity-to-Subject Constraint and the Type-of-Subject Constraint. Furthermore, this paper explores the force of the first constraint, which has recently been considered to be weaker than the second. Key words: Northern Subject Rule, Older Scots, Proximity-to-Subject Constraint, Type- of-Subject Constraint Resumen: Este trabajo versa sobre el funcionamiento de la llamada “Regla del norte” en la primera, segunda y tercera persona del plural del presente de indicativo en escocés antiguo. El estudio se basa en el análisis del corpus Linguistic Atlas of Older Scots y confirma que dicha regla opera en el presente de indicativo plural entre 1380 y 1500. El trabajo investiga cómo interactúan la restricción de Proximidad al Sujeto y la restricción de Tipo de Sujeto, al mismo tiempo que explora la operatividad de la primera restricción, que recientemente se ha considerado menos importante que la segunda. Palabras clave: Regla del norte, Escocés antiguo, Restricción de Proximidad al Sujeto, Restricción de Tipo de Sujeto 3 Contents 1. Introduction ............................................................................................................ 4 2. Brief outline of Older Scots ................................................................................... 4 3. Theoretical Background. The Northern Subject Rule ............................................ 6 3.1. Introduction ............................................................................................................. 6 3.2. Previous findings on the Northern Subject Rule ..................................................... 8 3.3. Previous findings on the Northern Subject Rule in Scots ..................................... 12 4. Research Questions and Hypotheses ................................................................... 15 5. Methodology ........................................................................................................ 16 6. Corpus Analysis ................................................................................................... 17 6.1. First person plural contexts ................................................................................... 17 6.2. Second person plural contexts ............................................................................... 23 6.3. Third person plural contexts .................................................................................. 25 7. Discussion ............................................................................................................ 36 7.1. First person plural contexts ................................................................................... 36 7.2. Second person plural contexts ............................................................................... 37 7.3. Third person plural contexts .................................................................................. 38 8. Conclusions .......................................................................................................... 39 References ........................................................................................................................ 42 APPENDICES ................................................................................................................. 47 Appendix I. First person plural syntactic contexts ....................................................... 47 Appendix II. Second person plural syntactic contexts ................................................. 48 Appendix III. Third person plural syntactic contexts ................................................... 49 4 1. Introduction The Northern Subject Rule is one of the most noteworthy features of Scots and northern Middle English. In broad terms, it is a grammatical constraint that governs present-tense verbal morphology (∅ vs. -s) according to the type and position of the subject (Fernández Cuesta, 2011: 90). Even though much research has been carried out recently, as Rodríguez Ledesma (2013) points out, there are still some issues to be discussed. Some of these matters are, for instance, the behaviour of the NSR in first person singular contexts, the development of the rule throughout the years in both English and Scots, or the predominance of one constraint over the other one. But it is precisely this last issue which concerns this study the most, as it focuses on the behaviour of the Northern Subject Rule in the present plural indicative in Older Scots (1380-1500), in an attempt to find out which of the two constraints seems to be more robust and in which specific contexts. The ultimate goal of the present study is to become part of the existing data on the operation of the Northern Subject Rule in Older Scots, which may be useful for other forthcoming studies. To achieve this, the paper begins with an introduction (section 1), followed by an outline of the historical development of Older Scots to provide essential context (section 2). Section 3 presents key information for the study, first explaining the operation of the NSR and then reviewing the existing literature, both from a broader perspective (section 3.2) and in more specific detail (section 3.3). Section 4 outlines the research questions and expected outcomes, while section 5 describes the methodology. The 'Corpus Analysis' (section 6) details the study's findings, which are then discussed in relation to the theoretical framework in section 7. Finally, the conclusion summarizes the key points and explores further implications of the research. 2. Brief outline of Older Scots Scots is a West Germanic language variety traditionally spoken in the Scottish Lowlands and northern islands. The number of Scots speakers varies depending on the source, but Millar (2023: 2) notes that, according to the 2011 census, there are approximately 1.5 million speakers, although the actual number is likely lower. The development of this linguistic variety is frequently linked to the history of the English language, as Scots “may be described through its processes of divergence from a shared ancestor, starting in the 11th century, and later, in the 16th-18th century, through the 5 convergence with English again during the period of anglicisation” (Gotthard, 2021: 61). Despite the close relation between English and Scots, Jones (2019: 7) denounces how the latter has sometimes been denied its own specific history, seen as divergent from English only in a few limited and easily defined areas. Farm from that, Scots is a variety with a strong identity, and it must be regarded as such, as it has its own distinctive elements and historical background. The history of Scots dates back to the 5th century, when the Angles, Saxons and Jutes, who spoke different dialects of the same Germanic language, settled in England (Rodríguez Ledesma, 1994: 2). The Angles expanded northwards, centering in Edinburgh, while attempting to control the British kingdom of Strathclyde to the west for centuries. The north-east was occupied by the Picts, who resisted the Romans and the Angles but were eventually unified with the Scots, forming the kingdom of Alba in the 6th century (Rodríguez Ledesma, 1994: 2). Corbett et al (2003: 5) explain that the Scots came from Ireland and spoke a Q-Celtic language, replacing the P-Celtic language spoken up to that moment.1 Rodríguez Ledesma (1994: 3) states that the Viking invasions in the 9th century influenced the language due to the similarities between Old Norse and Old English and hindered Anglo-Saxon control over Strathclyde and Lothian, which were ceded to Alba. By 1000, Scotland had two main languages: Gaelic in most parts and English in the southeast, with Norse in Shetland and Orkney. From 1100 onwards, an era of Scottish- English marriages kept up close relations between the countries (Mitchison, 2002: 13), and the Norman conquest in the 11th century introduced French and spread English due to the feudal system (Rodríguez Ledesma, 1994:3). The establishment of burghs bolstered English, and by the early 14th century, “Inglis” became official. Corbett et al (2003: 8) explain that: “Inglis superseded French in leisure and instructional reading and Latin in administrative prose; crucially, the Acts of the Parliament of Scotland began to be recorded in the vernacular in 1390, and the earlier Latin Acts were also translated in 1425”. Hence, by the late 14th century, Gaelic, dominant during the Norman rule, retreated to the Highlands, while “Inglis” became dominant in the Lowlands, linked to the feudal system and trade centres (Rodríguez Ledesma, 1994: 3). Finally, the years 1460 to 1560 can be regarded as “the heyday of the Scots tongue as a full national language” (Murrison, 1979: 8), as it showed all the signs of a rapidly developing, all-purpose speech. In short, 1 “The names ‘P’ and ‘Q’ Celtic refer to one variety having the sound `p' where the other has `q' (pronounced [kw] and usually spelled <c>).” (Corbett et al, 2003:5) 6 what it is now called Scots developed from an extended and complicated period of immigration and language contact. 3. Theoretical Background. The Northern Subject Rule 3.1. Introduction When one thinks about subject-verb agreement patterns, the first that comes to mind is the standard subject-verb English concord. Nonetheless, Ávila-Ledesma and Amador-Moreno (2023) find three other agreement types in different vernacular dialects: in some traditional dialects in the South-West of England and East Anglia, the verbal -s for third-person singular subjects is beginning to disappear, whereas in some southern and western dialects it has been generalised to all grammatical persons (Ávila-Ledesma and Amador-Moreno, 2023: 247). The third agreement pattern has been identified primarily in northern varieties such as Scots and North Midlands dialect, and it is the object of the present paper— a variable system that conditions verbal -s according to the type and position of the subject known as the Northern Subject Rule. This grammatical phenomenon is one of the most remarkable characteristics of the verbal morphology of Scots and northern Middle English, and it has been studied for years now, receiving different names depending on the scholar: McIntosh (1989: 117) also refers to the NSR as “the northern paradigm”. The term “Northern Subject Rule” was coined by Ihalainen (1994: 221), but the constraint is also referred to as the “Northern Present-Tense Rule” (Montgomery, 1994: 83), “they-constraint” (Wright 2002: 243) and “northern concord rule” (García-Bermejo Giner and Montgomery 2003: xxxiii). (Cole, 2008: 92) But whatever the name used, the important issue is that, according to the NSR, “present-tense indicative verb endings are conditioned not just by person and number of the subject, but also, intriguingly, by its word category (personal pronoun or nominal) and its position relative to the finite verb (adjacent or non-adjacent to it)” (de Haas and van Kemenade, 2015: 49). This means that the verbal form will typically take a zero ending (-Ø) if the subject is a personal pronoun and is adjacent to the verb. On the contrary, when 7 the subject is nominal or there are elements between the subject and the verb, the verb will show the inflection -s, as de Haas and van Kemenade (2015: 50) illustrate: (1) (a) they sing [Spro - V-Ø] (b) birds sings [SNP - V-s] (c) they sing and dances [Spro - . . . - V-s] (d) they always sings [Spro - . . . - V-s] The NSR is traditionally explained by distinguishing between its two different constraints: the Type-of-Subject Constraint and the Proximity-to-Subject Constraint. According to Montgomery (1994: 88), the Type-of Subject Constraint affects the verb “if its subject is anything but an adjacent personal pronoun”, whereas the Proximity-to- Subject Constraint affects “any verb having a personal pronoun subject not adjacent to the verb”. Rodríguez Ledesma (2013) follows Macafee (2002), who explains that if the subject is a personal pronoun (Type-of-Subject Constraint) and is adjacent to the verb (Proximity-to-Subject Constraint), the inflections are the ones shown in picture 1. If that is not the case, there will be -s for all persons and numbers, as represented in picture 2. Macafee (2002: cxii) adds that in those cases where the personal pronoun governs two adjacent verbs, the first one is affected by the contiguity, so it will have zero inflection, whereas the second will have -s. Picture 1 (Rodríguez Ledesma, 2013: 150) Picture 2 Rodríguez Ledesma (2013: 150) Let us see how these two constraints work with the following examples taken from Cole (2014: 35-36): (2) Thay kepe thaire wynges clene (The Bee and the Stork, Thornton, Lincoln Cath. 91 [f. 194]) ‘They keep their wings clean’ 8 (3) Swa þay hafe vndirstandynge, and fastes and wakes and semes haly to mens syghte (The Bee and the Stork, Thornton, Lincoln Cath. 91 [f. 194]) ‘So they have understanding and fast and wake and seem holy to men’s sight’ (4) Twa thynges makes oure delyte pure (Desyre and Delit, Thornton, Lincoln Cath. 91 [f. 196b]) ‘Two things make our delight pure’ (5) Thus es it of thayme þat turnes þame to godes seruys (The Bee and the Stork, Thornton, Lincoln Cath. 91 [f. 194]) ‘Thus, it is of them that turn themselves to God’s service’ As shown in the verbs kepe and hafe from examples (2) and (3) respectively, when the pronominal subject (thay) is adjacent to the verb, the verb has zero inflection. On the contrary, when a pronominal subject is not adjacent to the verb, the verb has the -s ending as seen in fastes, wakes and semes in example (3). The inflection -s is also found in examples (4) and (5), where the subject, although adjacent to the verb, is not a personal pronoun, but a noun phrase (twa thynges) or a relative (þat). In summary, these examples show the behaviour of the NSR rule. 3.2. Previous findings on the Northern Subject Rule The NSR has aroused interest among linguists for decades now, and there is a broad consensus within historical linguistics that it began operating in Middle English. However, there has been some research about it in other language varieties, and some recent studies even propose an early appearance of the NSR; for instance, Cole (2011) suggests that there is a tendency for subject type and adjacency to condition verbal morphology in Old Northumbrian. She carries out a study of the verbal morphology of the Lindisfarne Gospels (glossed in Old Northumbrian around 950) and challenges the traditional view by stating the following: It has tacitly been assumed in the literature that the reduced forms at the crux of the NSR, and the constraint that triggers them, must have emerged in the northern dialects during the early Middle English period, as there is little indication of the 9 pattern existing in extant Northumbrian texts from the tenth century, and by the time northern textual evidence is once again available from c.1300, the NSR is clearly prevalent (Pietsch 2005; de Haas 2008; de Haas & van Kemenade 2010). Nevertheless, the assumption that the NSR was entirely lacking in Old Northumbrian stands on shaky grounds without further detailed analysis of the tenth-century northern writings as has been pointed out in the literature (Benskin 2011: 170). (Cole, 2011: 9) According to Kiriko’s (2017) review on Cole’s research, she uses two main pieces of evidence to support the hypothesis that the NSR may have started in Old Northumbrian. First, she explains that the reduced verb endings -e/-o/-a might have been caused by the NSR, as is the case in northern Middle English as well. Although this hypothesis has raised doubts, since these endings are rare (found only 7.6% of the time), Cole (2011: 117) suggests that because earlier research only looked at personal pronouns, a broader study that includes other types of subjects might show different results. Secondly, the choice between endings like -s and -ð or -s and -e/-Ø might also be influenced by the NSR in Old Northumbrian, as this rule does not just separate inflected and uninflected forms, but instead involves choosing between different forms depending on the sentence structure (Kiriko, 2017). With regard to the origin of this syntactic constraint, there are mainly two opposing views according to de Hass (2006). On the one hand, there is a theory posed by Hamp (1975) and held by Klemola (2000) that attributes the NSR to the effects of language contact with Celtic. This view would align with Cole’s, as it argues that the NSR may be an underlying result of Brythonic Celtic, originating in a period of Cumbrian/Anglo-Saxon bilingualism in Northumbria (de Hass, 2006: 112). On the other hand, those that are more skeptical of the influence of language contact, attribute the appearance of the NSR to language-internal effects (Pietsch, 2005). De Haas (2011: 216, 217) explains that these language-internal hypotheses for the development of the NSR suggest that the pattern emerged through the generalization of -e/Ø verb endings, likely stemming from inversion contexts with plural pronoun subjects in Old English, where the indicative already featured -e. This was eventually combined with the generalized plural -s form (de Haas, 2011: 217). In spite of the various hypotheses and extensive linguistic research, the exact origins of the NSR remain elusive, leaving it an ongoing subject of scholarly debate. 16 5. Methodology The present paper carries out a corpus study on the behaviour of the Northern Subject Rule in the present plural indicative in Older Scots (1380-1500) using the online corpus LAOS, more specifically version 1.1, 2008-2013, and version 1.2, 2013. It was created by the researcher Keith Williamson from the University of Edinburgh. This corpus aims to present information about the spatial and temporal variation of linguistic forms found in Older Scots texts. Older Scots conventionally covers the period 1150 to 1700, although extant continuous texts date between 1380 and 1700. This first phase of the atlas mainly contains 'local documents', that is, writings localized from internal, non-linguistic references, and covers the earlier period, 1380 to 1500. This was one of the main reasons for choosing this corpus, as it includes the earliest records of the language. In addition, the records of texts are based on diplomatic editions, transcribed directly from manuscripts or facsimiles, which makes it possible to access the most reliable data (Rodríguez Ledesma, 2013: 158). LAOS is a ‘daughter’ atlas of A Linguistic Atlas of Late Mediaeval English, 1350- 1450 (LALME), and it expands the limited coverage of Scots in LALME to treat in greater depth texts from the entire geographical area where Scots was used. Furthermore, LAOS is also related to A Linguistic Atlas of Early Middle English, 1150-1325 (LAEME), which deals with variation in written English before LALME's period. Both LALME and LAEME have been consulted for this paper to find out the meaning of some abbreviations or conventions used in LAOS and not found in the introduction to that corpus6. In addition, when there was any difficulty regarding the understanding of lexical items, the online version of the Dictionaries of the Scots Language (DSL) was used, selecting the filter option ‘Before 1700’. This online tool was also useful for the translations of the examples. Once explained the tools employed in the study, the procedures that have been carried out must also be described. First and foremost, it should be highlighted that this study is based on a quantitative analysis to avoid any kind of bias. To begin with, the analysis is divided by person, i.e., the first person plural contexts are first analysed, then the second person plural and finally the third person plural, repeating the same process of analysis for each one. To do so, a search for the tag ‘vps2’ (plural present indicative) is 6 An update of LAOS was in the process of being carried out, which may be the reason why this information was not available in the online version of the corpus. 17 carried out using the task ‘Concordances’ and selecting the option ‘inflexions’. The task divides the results into different syntactic contexts. As mentioned before, this study has focused on the ones corresponding to ‘vps21’, ‘vps22’ or ‘vps23’ for the first, second and third person, respectively, obtaining a total number of 987 tokens. The task also differentiates among the different syntactic contexts within each person. In order to simplify the analysis, these syntactic contexts have been regrouped according to their similarities, and some minority contexts have been disregarded, as there is a low number of examples, or they are not representative. For instance, in the firstperson plural, the search returned 23 different syntactic contexts, which have been grouped in three main syntactic contexts and three minor ones. The different groupings for each person will be seen in section 6. After that, the results of the different contexts have been classified according to whether they present the inflection -s or not. Even though there are different spellings for the inflection (-is, -iss, -ce, -es, -ys, -z), these are considered as realisations of the same verbal ending. Once the general results have been analysed, some exceptions and specific examples have been discussed when necessary. 6. Corpus Analysis 6.1. First person plural contexts The search for the tag ‘vps21’ displays a list of 23 different syntactic contexts7 and a total number of 272 examples. They have been grouped into the six major categories below: (a) /vps21<P+, /vps21<P&P+ and /vps21<P^P+: The subject pronoun is adjacent to the verb, or two subject pronouns coordinated are adjacent to the verb. Coordination occurs either by means of the conjunction and (&) or or (^). (13) and ferder we ordane thame to causs. (/vps21<P+) ‘And further we ordain thame to cause.’ (Edinburgh, National Library of Scotland Ch 13024) 7 See Appendix I for the complete list of contexts. 18 (14) j and he raman in sic-lik richtis as we war jn of-befor. (/vps21<P&P+) ‘I and he remain in suchlike right as we were in before.’ (Edinburgh, National Library of Scotland, Ch. 15573) (15) j or ya remanis w^t ye said landis off Balbertane. (/vps21<P^P+) ‘I or you remain with the said lands of Balbertane.’ (Edinburgh, National Library of Scotland MS 73, item 5) (b) /vps21<P-, vps21<P^P- and /vps21<P-<=: The subject pronoun or two subject pronouns coordinated by the conjunction or are not adjacent to the verb. The last tag (/vps21<P-<=) refers to a construction formed by a phrasal verb. (16) we forsaid jugis approviss and ratifies ye sentence diffinitife geffyne be master Simond of dalgless. (vps21<P-) ‘We, the aforementioned judges, approve and ratify the definitive sentence given by master Simon of Dalgles.’ (Edinburgh, National Archives of Scotland, HM General Register House GD28/157) (17) & gefe j or ya yus warnyt comperis no^t ye said day and place and resafis no^t our said sowm of mone. (vps21<P^P-) ‘and if I or you, thus warned, don’t compear the said day and place, and don’t receive our said sum of money.’ (Edinburgh University Library, Laing Charters 1269, box 34) (18) & here we ye forsaid walter and thomas procuratoris to ye said robert for hym and in his name falss it adnullis it and agane callis it in yi hande. (/vps21<P-<=) ‘and here, we (the aforementioned Walter and Tomas, procurators to the said Robert) for him and in his false name it annuls it and call again it your hand. (Edinburgh, National Library of Scotland Ch 15552) (c) /vps21<PO-: The subject is the object pronoun (us) and is not adjacent to the verb. 19 (19) us dauid be ye permissione off gode abbot of arbroth ande ye conuent off ye sammyn till haue set ande to ferme lattyn & be ye tenore off yir lettres settis & to ferme lattis till oure louet johne alan sone burgesss off dunde till margret & katrine hyss dochteris. ‘Us, David, with the permission of the good abbot of Arbroath and the convent thereof, to have set and to have leased the land by the terms of these letters, set and lease the land to our beloved John Alan, son of a burgess of Dundee, to Margaret and Katrina, his daughters.’ (Edinburgh, National Library of Scotland Adv. MS 34.4.3, f. 114r, item 357) (d) /vps21<P&n+ and /vps21Pn&n+: The subject is a personal pronoun coordinated with a noun (‘I/me/we and X’) adjacent to the verb. Other variants within this group are: /vps21<P^n+, /vps21<Pn^n+, /vps21<Pn^n+>=. (20) we & ye said place sustenis in ye persut of diuerss vrangis dwne to ws. (/vps21<P&n+) ‘We and the said place support in the pursuit of different wrongs done to us.’ (Edinburgh, National Library of Scotland Adv. MS 34.4.3, f. 79 - f. 80v, item 255) (21) j my ayris executouris & assignais grantis ws ful dettouris als weil as ye principal sowm. (/vps21Pn&n+) ‘I, my heirs, executors and assignees grant us full debtors as well as the principal sum.’ (Edinburgh, National Library of Scotland Adv. MS 34.4.3, f. 63v - f. 64r, item 220) (e) /vps21<P&n-, /vps21<Pn&n- and /vps21<Pn&n->=: The subject is a personal pronoun coordinated with a noun (‘I/me/we and X’) not adjacent to the verb. The last tag (/vps21<Pn&n->=) refers to a phrasal verb. Other variants within this group are /vps21<P<^n^n-, /vps21<P^n-, /vps21<P^n^-, /vps21<P^n^n-, vps21<P^nn^n-<= and /vps21<Pnn^n-. 20 (22) me & walter-home my sone makis jnterrupcione to ye pretendit stat. (/vps21<P&n-) ‘Me and Walter-Home, my son, make interruptions to the pretended state.’ (Edinburgh, National Archives of Scotland, HM General Register House B22/22/4, f. 21v) (23) j myn heiris to be excludy al-vtralye fa ye said landis w^t yair pertinence and nowe as yan & yan as nowe renunsis quytclamys & giffis our al richtis clamys & titile of richtis. (/vps21<Pn&n-) ‘I and my heirs, to be completely excluded from the said land with their property, and now as then and then as now, renounce, quitclaim and give over all rights, claims and title of rights.’ (Edinburgh, National Archives of Scotland, HM General Register House GD220/20/158) (24) j myn heiris […] giffis our al richtis clamys & titile of richtis. (/vps21<Pn&n->=) ‘I and my heirs […] give over all rights, claims and title of rights.’ (Edinburgh, National Archives of Scotland, HM General Register House GD220/20/158) (f) /vps21<n- and /vps21<n&n-: The subject is a noun or two coordinated nouns, respectively, not adjacent to the verb. (25) ye present lettres in owr common buk settis. (/vps21<n-) ‘The present letters lay in our common book.’ (Edinburgh, National Archives of Scotland, HM General Register House B58/8/1, f. 119r - f. 122v (7)) (26) wylzam dickyson & john morthoson balzeis of y^e burgh of peblis w^t consent & assent of y^e hayll communete of y^e said burgh, grantis ws tyll haf setting. (/vps21<n&n-) 21 ‘William Dickinson and John Morthoson, bailies of the burgh of Peblis, with the consent and assent of the whole community of the said burgh, grant us to have setting.’ (Edinburgh, National Archives of Scotland, HM General Register House B58/8/1, f. 119r - f. 122v (7)) The results obtained in the automatic search have been classified according to these contexts, as shown in Table 2 below:8 Table 2. LAOS results for 1pl.pres.ind n∅ n-s N Tot % ∅ % -s (a) Subject pronoun(s) adjacent to the verb 95 3 98 96.94 3.06 (b) Subject pronoun not adjacent to the verb 1 122 123 0.81 99.19 (c) Object pronoun not adjacent to the verb 0 18 18 0 100 (d) Subject pronoun coordinated with a noun adjacent to the verb 1 12 13 7.69 92.31 (e) Subject pronoun coordinated with a noun not adjacent to the verb 0 17 17 0 100 (f) Subject noun(s) not adjacent to the verb 0 3 3 0 100 The results of contexts (a), (b) and (c) show that in the majority of the cases where the subject pronoun or pronouns are adjacent to the verb, no inflection is present; on the contrary, the inflection is found when the subject pronoun and the verb are not contiguous in accordance with the NSR. This means that there is a near categorical operation of the Proximity-to-Subject Constraint in these specific contexts. There are only four exceptions to this phenomenon, so they will be looked into more closely: (27) to the-quhilkis the said parteis ar oblist til hald & thai decetis & ordains apon a debate of land. ‘To which the said parts are obliged to hold, and they decreet and ordain upon a dispute of land.’ 8 The Result tables are based on the charts from Rodríguez Ledesma (2013). 22 (Aberdeen Council Register II, 21 Jan 1453/54, p194) In example (27), the verb decetis has the inflection -is, even though it is placed immediately after the subject pronoun. However, the subject pronoun is the third person plural (thai), so it can be assumed that it must be an error in the webpage code (vps21<P+ instead of vps23<P+). Therefore, this specific example has been dismissed. Let us see the other three exceptions: (28) thus we deliueris of ane assent. ‘Thus, we deliver one sanction.’ (Edinburgh, National Archives of Scotland, HM General Register House GD28/157) (29) quhilk name it gat as we weyll knaw in defraude of ws. ‘which name it got as we clearly know in order to deceive us.’ (Edinburgh, National Library of Scotland Adv. MS 34.4.3, f. 79 - f. 80v, item 255) (30) for als lang zeris & termis as j or ya remanis w^t ye said landis off balbertane. ‘for as long years and terms as I or you remain with the said lands of Balbertane.’ (Edinburgh, National Library of Scotland MS 73, item 5). In example (28), the underlined verb has the inflection -s despite being adjacent to the subject pronoun, whereas the verb knaw in example (29) has zero inflection despite being not adjacent to the subject. In example (30), the subject pronouns are coordinated by the conjunction or and are adjacent to the verb, but the verb has the inflection -is. Although the verb is tagged as first person plural, because of the conjunction or it could be considered as agreeing with the second person singular pronoun ya, which requires -s. On the other hand, in the contexts where the subject is the object pronoun us and is not adjacent to the verb, all the cases show the use of the inflection according to the NSR, showing a categorical operation of the Proximity-to-Subject Constraint. Whereas contexts (a), (b) and (c) illustrate the Proximity-to-Subject Constraint, (d), (e) and (f) exemplify the Type-of-Subject Constraint. The results from Table 2 show a near-categorical operation of this constraint, since in those cases in which the subject is 23 a pronoun coordinated with a noun or a combination of nouns, the verb takes the inflection -s regardless of adjacency. There is just one instance that does not conform to the rule: (31) qwyllys we or oure ayerrys mak tyll forsayde syre harbarte. ‘while we or our heirs make to foresaid sir Harbarte.’ (Hull, University of Hull, Brynmor Jones Library DDEV/80/7) It is interesting to note that in (31) the subject pronoun is coordinated with a NP by means of the conjunction or (cf. example (30) above). However, in similar contexts the verb does take the inflection -s. (32) we ore oure ayerrys hyghtys to mak ore gare be made to y^e forsaid syre harbarte. ‘We or our heirs promise to make or cause to be made to the foresaid sir Harbarte.’ (Hull, University of Hull, Brynmor Jones Library DDEV/80/7) 6.2. Second person plural contexts In the case of the second person plural, a search for the tag ‘vps22’ was conducted. In this case, the task returned 18 examples divided in four syntactic contexts,9 which have been grouped together in a similar way as before: (a) /vps22<P+: The subject pronoun is adjacent to the verb. (33) again yhe suffer to be shippit & frely to passs w^t-outyn ony addyng. ‘Again you suffer to be sent and free to pass without any adding.’ (Edinburgh, HM Register House, SRO GD 55/480: 26 May 1389) (b) /vps22<P- and /vps22<P->=: The subject pronoun is not adjacent to the verb. The second tag (/vps22<P-<=) refers to a construction formed by a phrasal verb. 9 See Appendix II for the complete list of syntactic contexts. 24 (34) yat yhe compere in ye said place ye last day of ye mone^t^h of october next-to-cum wy^t continua\cion of dayis & endis (/vps22<P-) & giffis furth (/vps22<P->=) ye said submission. ‘That you appear in the said place, the last day of the month of October next to come with the continuation of days and end and give forth the said submission’. (Edinburgh, National Archives of Scotland, HM General Register House GD26/1/4/43) (c) vps22<P^n+: The subject is a personal pronoun coordinated with a noun adjacent to the verb. (35) In ye meyin-tyme ze or zoure convent be zour-self or officiaries dois hir ony trubile. ‘In the meantime, you or your convent by yourself or officers do her any trouble.’ (Facsimile: Edinburgh University Library Ph. 1690, whence transcribed) The results for these syntactic contexts are displayed in Table 3 below. As can be seen, there is a categorical operation of the Proximity-to-Subject Constraint: in all the cases where the subject pronoun is adjacent to the verb, it has zero inflection, whereas -s occurs in those contexts where there is distance between the subject pronoun and the verb. In addition, the Type-of-Subject Constraint is illustrated in context (c): when the subject is a pronoun coordinated with a noun, the verb takes the inflection -s. Table 3. LAOS results for 2pl.pres.ind n∅ n-s N Tot % ∅ % -s (a) Subject pronoun adjacent to the verb 15 0 15 100 0 (b) Subject pronoun not adjacent to the verb 0 2 2 0 100 (c) Subject pronoun and a noun adjacent to the verb 0 1 1 0 100 25 6.3. Third person plural contexts For the analysis of the third person plural contexts, a search for the tag ‘vps23’ returned a list of 39 different syntactic contexts and a total of 695 tokens.10 These contexts have been divided into four categories with different subgroups: (a) Personal pronoun as subject: a) /vps23<P+, /vps23<P&P+, /vps23<P^P+ and /vps23<P+>=: The subject pronoun is adjacent to the verb, or two subject pronouns coordinated are adjacent to the verb. Coordination occurs either by means of the conjunction and (&) or or (^). The last tag (/vps23<P+<=) refers to a construction formed by a phrasal verb. (36) on t the-todir part yat yai ar fullely acordit. (/vps23<P+) ‘On the other part, that they are fully in agreement.’ (Edinburgh, National Library of Scotland Ch 15560) (37) qwhill he and j apper jn personis be-for ye chapatire. (/vps23<P&P+) ‘While he and I appear in person before the chapter.’ (Durham, Prior's Kitchen, Dean and Chapter Muniments: Locellus XXV.6) (38) na thing yat he ore yai resayvis of ye saide towne. (/vps23<P^P+) ‘No thing that he or they receive of the said town.’ (Durham, Prior's Kitchen, Dean and Chapter Muniments: Misc. Charter 1162) (39) yat sal be gewyn & payt or yai com out of y^e tolbuth. (/vps23<P+<=) ‘That shall be given and paid before they come out of the tolbooth.’ (Edinburgh, National Archives of Scotland, HM General Register House B58/8/1, f. 73r - f. 81v (21)) 10 See Appendix III for the complete list of contexts. 32 (60) and gif it hapnis y^t y^e saide ser williame-more or rankyn his son or ony of yair ayris or ony of yaire successouris or ony oyer on yaire behalf makis claime or chalang in tyme to cum in y^t superiorite to yat lande. ‘And if it happens that the said sir William More or Rankyn, his son, or any of their heirs, or any of their successors, or any other on their behalf, make claim or challenge in time to come in that superiority to that land.’ (Edinburgh, National Library of Scotland MS 72, f. 22r-v) k) /vps23<P^n+ and /vps23<n&P+ (3 cases): The subject is a personal pronoun coordinated with a noun by means of and or or and adjacent to the verb. (61) & swa ye place & we sustenis yir vrangys in yir said landis. ‘And so the place and we sustain these wrongs in these said lands.’ (Edinburgh, National Library of Scotland Adv. MS 34.4.3, f. 79 - f. 80v, item 255) l) /vps23<P^n-, /vps23<P^n^n-, /vps23<Pn^n- and /vps23<Pn^n->= (4 cases): The subject is a personal pronoun coordinated with a noun or several nouns and not adjacent to the verb. The last tag (/vps23<Pn^n->=) refers to a construction made up by a phrasal verb. (62) & at ye will of ye for-said lordis be-for wrytyn yar ayris executuris & assigneis to be sauld til als wele yar costis expenss skathis & jnteresss gyf yai or ony of yaris sustenys in defalt of pament. ‘And at the will of the foresaid lords before written, their heirs, executors and assignees; to be sold to as well their costs, expenses, scathes and interests, if they or any of them sustain in default of payment.’ (Edinburgh, National Archives of Scotland, HM General Register House RH1/6/65) The tables below present the results for the different syntactic contexts. In LAOS there are 110 instances of category a) in which the subject is a personal pronoun. Table 4 provides us with the figures depending on whether there is adjacency or not: 33 Table 4. LAOS results for 3pl.pres.ind. Personal pronoun as subject n∅ n-s N Tot % ∅ % -s a) Subject pronoun(s) adjacent to the verb 88 5 93 94.62 5.38 b) Subject pronoun not adjacent to the verb 1 15 16 6.25 93.75 As may be noticed, the percentages are similar to the ones for the first person plural. In the first syntactic context, when the subject pronoun and the verb are adjacent, the verb has zero inflection in 94.62% of the cases. On the contrary, in the second context, when they are separated from each other, the verb takes -s in 93.75% of the cases in accordance with the NSR. It is interesting to notice that the five examples that do not conform to the NSR in group a) are those where the conjunction or links two subject pronouns, as can be seen in (38). So, this conjunction seems to trigger the inflection -s regardless of the type and the position of the subject, as illustrated by the additional examples below: (63) as he ore thai thynkis expedi {=hole 5 or 6 words lost=} ‘As he or they think expedient’ (Edinburgh, National Archives of Scotland, HM General Register House GD99/4/1) (64) y^t he or thai sustenis in ye non-peciabile brouking ‘that he or they sustain in the not peaceful possession. (Edinburgh, National Archives of Scotland, HM General Register House GD99/4/1) (65) has he or yai wantis & omyttis for al ye termes & tymis at ye said landis. ‘as he or they want and omit for all the terms and times at the said lands.’ (Edinburgh, National Archives of Scotland, HM General Register House RH6/395) (66) a toure & fortalice quhare he or thai thinkis maist expedient. ‘a tower and fortalice where he or they think most expedient.’ 34 (Edinburgh, National Archives of Scotland, HM General Register House RH1/6/162) There are 203 instances of category b) in which the subject is not a personal pronoun but a relative. Table 5 below shows the results for this context: Table 5. LAOS results for 3pl.pres.ind. Relative as subject n∅ n-s N Tot % ∅ % -s (d) Subject is a relative adjacent to the verb 1 96 97 1.03 98.97 (e) Subject is a relative not adjacent to the verb 0 106 106 0 100 Contexts (d) and (e) illustrate the Type-of-Subject Constraint: when the subject is not a personal pronoun, the verb takes the inflection -s regardless of adjacency. In the remaining 337 instances the subject is a plural noun phrase or a combination of nouns (category c), as can be seen in Table 6: Table 6. LAOS results for 3pl.pres.ind. Plural NPs as subject n∅ n-s N Tot % ∅ % -s (f) Subject noun(s) adjacent to the verb 0 143 143 0 100 (g) Subject noun(s) not adjacent to the verb 4 190 194 2.06 97.94 Here it can be observed that the findings are similar to those in Table 5 and illustrate the Type-of-Subject Constraint: whatever the distance between the subject and the verb, the verb shows -s in accordance with the NSR. The only exceptions are the following four examples: (67) as ye jndentours made betwixt yaim ber witness yareof ‘as the indentures made between them bear witness thereof.’ (Edinburgh, National Archives of Scotland, HM General Register House GD55/514**) 35 (68) y^e forsaide sir williame his ayris and his assignez for me and myne ayris and myne executouris and assignez euer-mare awytcleme y^e forsaide landis to halde. ‘the foresaid sir William, his heirs and his assignees, for me and my heirs and executors and assignees, evermore quitclaim the foresaid lands to hold.’ (69) sen gode off his grace hes giffyn ye place largely to luff apon & yat sleuithness off officiaris ger ye place want prouisione. ‘Since then, God of his grace has given the place largely to live upon and that slothness of officers cause the place to want provision.’ (Edinburgh, National Library of Scotland Adv. MS 34.4.3, f. 121v - f. 122r, items 390 – 39) (70) the wardanis on al ye marchis ger set and hald wardane courtis ‘The guards, on all the borders, force to set and hold the courts of the guards.’ (Edinburgh, National Archives of Scotland, W Register House, PA 2/2, f61r - f64v) In order to be in accordance with the NSR, the verbs in these examples should take the inflection. In (67) it can be argued that there is a personal pronoun attached to the verb, and although it is not the subject, it might have led to confusion. It is interesting that (69) and (70) are two examples of the same verb (gar), which may suggest that this verb typically appears uninflected. However, the DSL also gives examples of third person singular and plural with inflection. Thus, there is no sufficient explanation for the exceptions, which shows that even though the NSR has proven to be near-categorical in the corpus analysed, linguistic variation may occur at any time in any utterance. 36 7. Discussion 7.1. First person plural contexts The analysis of the 272 instances of this context allows to test the operation of both constraints, even though most of the examples illustrate the Proximity-to-Subject Constraint. According to the recollected data, when the subject pronoun or pronouns are adjacent to the verb, there is a tendency for zero inflection (96.94%), whereas the inflection is found when the subject pronoun and the verb are not contiguous (99.19%). When the subject is realised by an object pronoun not adjacent to the verb, it takes the inflection as well (100%). This means that adjacency plays a vital role in the analysed corpus; thus, there is a near-categorical operation of the Proximity-to-Subject Constraint in the first person plural. Montgomery (1994: 89) reaches a similar conclusion in his study, finding that in contexts where the subjects are non-adjacent personal pronouns, 106 out of 113 verbs show the inflection (94%). Even though 41 of these cases belong to fixed phrases at the end of letters or correspond to contexts where there is a null subject, Montgomery (1994: 93) confirms the near-categorical operation of the Proximity-to-Subject Constraint in the first person plural. This is supported by Rodríguez Ledesma’s (2017: 817) findings, as she noted that in 151 out of 155 examples of her corpus, the verb took zero or -s in accordance with the NSR (97.4%). Although some of the exceptions to the application of the Proximity-to-Subject Constraint in the first person plural have been explained, no justification could be found for some others. These exceptions may be coincidental, showing that the rules of language are not fixed, and that linguistic variation may occur at any time in any utterance. Another possible explanation for those exceptions is that there are verbs which behave differently due to their own nature. Rodríguez Ledesma (2013: 169) explains in her study the behaviour of be and have: The data from LAOS confirm earlier statements about the behaviour of different verbs with regard to the operation of the NSR (Murray 1873, Kuipers 1964, Montgomery 1994, among others): in the case of be the NSR operates as a variable rule (it applies optionally and inconsistently), while have patterns with lexical verbs and follows the NSR practically categorically. 37 However, this explanation does not fully account for the exceptions observed in the corpus of this study. Since no consistent pattern or recurring verb that breaks the rule can be identified, the only possible conclusion is that these deviations are purely coincidental. Something that is worth mentioning is the exception in example (15), where the two subject pronouns are adjacent to the verb. In this case, according to the NSR, the verb should have zero inflection; however, this example does not conform to the rule, as it has the inflection -s. The only difference with the other examples is that the pronouns are coordinated by means of the conjunction or, which might suggest that this conjunction triggers the use of the inflection. This hypothesis is reinforced by similar examples in other syntactic contexts, as will be seen in the third person plural. With regard to the cases where the subject is realised by the object pronoun, the LAOS data indicate that the oblique pronoun us functions similarly to the nominative we in relation to the application of the NSR, just as Rodríguez Ledesma (2013: 169) finds in her corpus. In LAOS, it always occurs separated from the verb, and always takes an -s suffix, following categorically the Proximity-to-Subject Constraint. On the other hand, the Type-of-Subject Constraint, despite being less frequent in the first person plural, also operates near-categorically: out of the 33 cases, 32 present the inflection -s in the verb (97.43%). The results are similar to those found by Rodríguez Ledesma (2017: 812) in her study of the Breadalbane Collection: “With regard to the Type-of-Subject Constraint, the NSR is found to be categorical as well, since in all the examples in which there is no explicit or zero pronoun (18 tokens in the corpus) the verb takes -s”. The only exception found in our corpus seems to be once again the coordination by means of the conjunction or (see example (31)). In spite of that, in summary, the results for the first person plural reveal that both constraints are fully operative and that they are equally strong. 7.2. Second person plural contexts Although there are far fewer examples for the second person plural, the results are even clearer than in the first person plural. There is a categorical operation of both the Proximity-to-Subject Constraint and the Type-of-Subject Constraint, which means that they are both operative in the corpus and equally robust. However, the results may be 38 unbalanced, as there are only 17 instances of the former and one of the latter in the analysed corpus. Even though this is expected in these contexts, as it is difficult to find written documents directed towards the reader (especially legal and formal ones), it reduces the chances of having variation in the rule. The existent research has frequently encountered this problem of lack of data for the second person. For instance, Gotthard (2021: 65) notes that her analysis of the second-person pronoun was insufficient because she assumed that both singular and plural forms are affected by the NSR. However, she did not include any examples of second-person plural referents, which meant she may have overlooked whether a distinction based on number actually exists (Gotthard, 2021: 65). Even Montgomery (1994: 89) faces this problem; despite obtaining the same results as the present study— a categorical operation of both constraints for the second person plural— he could only find three instances in his corpus. Likewise, Rodríguez Ledesma (2017: 817) explains that very few examples are attested in non-adjacent position in the Breadalbane Collection (only 6 tokens), and that there are no examples to test the Type-of-Subject Constraint. Nonetheless, by means of those few contexts, she could prove the strength of the Proximity-to-Subject Constraint. The findings of the present study, therefore, corroborate those results, demonstrating that there is a categorical operation of both constraints in the second person plural. 7.3. Third person plural contexts Contrary to the second person plural, the discussion of the third one is much more intricate. The complexity arises primarily from the variety of subjects that third person plural can take, a necessity stemming from their frequent use in legal texts such as the ones found in LAOS. As a result, the operation of both NSR constraints is tested, especially the Type-of-Subject Constraint. The results for the third person plural are as expected according to the NSR; adjacency is key when the subject is a personal pronoun, whereas it is the type of subject that determines the use of the inflection when the subject is a pronoun coordinated with a noun, a noun phrase or a relative. In other words, in the examples where the Type-of- Subject Constraint was tested, 99.23% conformed to it, while when the Proximity-to- Subject Constraint was studied, 96.12% of the examples behave accordingly. This shows that both constraints are not just operative but also nearly categorical in the third person 39 plural. This verifies previous research like Meurman-Solin’s, who found that “with plural noun subjects, the suffixless predicate verb is rare” (1993: 619). In addition, Rodríguez Ledesma (2017: 815), despite not finding instances to test the Proximity-to-Subject Constraint, reveals that the Type-of-Subject Constraint is near-categorical (the verb adds -0 when adjacent to a personal pronoun (28 out of 29 tokens, 96.5%), and -s when the subject is not a pronoun (35 out of 38 tokens, 92.1%). These findings, along with the ones of the present study, align with Montgomery’s research: For the other side of the type of subject constraint, there is only one occurrence in 112 contacts throughout the data (a rate of 0.9%) of -s on a verb having an adjacent they as subject, an instance that cannot be directly explained by the encroachment of Anglicization. Otherwise, there is no evidence that verbs with an adjacent they (or any other personal pronoun, except second person and 3rd personal singular contexts) are ever marked with -s. (Montgomery, 1994: 87) The construction with two personal pronouns connected by the conjunction or exhibits deviations from the rule, as seen in the first person plural. In the five examples in which the subject pronouns are coordinated by or and are adjacent to the verb, the verb takes the inflection -s (cf. instances (38) and (63)-(66)). This inflectional behaviour contrasts with situations where the subject pronouns are connected by and, revealing a significant difference in how conjunctions affect verb agreement. 8. Conclusions The present study confirms that the plural forms of the present indicative were subject to the NSR between 1380 and 1500 based on the corpus analysis of the Linguistic Atlas of Older Scots. Previous studies had shown that the NSR was operative in the present indicative plural in Older Scots, and this paper provides additional data that show that it was indeed very productive in first, second and third person plural. The main goal was to examine if both constraints were operative and equally robust in the analysed corpus. After a careful analysis, the study indicates that, contrary to what happens in northern Middle English and the statements found in previous literature, such as de Hass (2006), the Proximity to Subject Constraint is not weaker than 40 the Type of Subject Constraint in Older Scots. On the contrary, in the analysed corpus of LAOS (1380-1500), the Proximity-to-Subject Constraint is as strong and categorical as the Type of Subject Constraint, as illustrated in Table 7: Table 7. LAOS results for the NSR in present indicative plural Type of Subject Constraint Proximity to Subject Constraint First person plural Near-categorical (97.43%) Near-categorical (98.7%) Second person plural Categorical Categorical Third person plural Near-categorical (99.23%) Near-categorical (96.12%) These findings demonstrate that the Proximity-to-Subject Constraint operated almost categorically: out of a total of 141 examples in non-adjacent contexts in the corpus, 139 (98.58 %) take a sibilant ending. These results confirm the hypothesis of this study and align with earlier research by Meurman-Solin (1993), Montgomery (1994), and Rodríguez Ledesma (2013, 2017). For instance, in her study on the operation of the NSR in the Breadalbane Collection, Rodríguez Ledesma (2017: 821) concludes: The study also reveals that, in contrast with northern Middle English, where the Proximity to Subject Constraint has been found to be much less robust than the Type of Subject Constraint, in the Breadalbane Letters it is near-categorical and as strong as the Type of Subject Constraint. These findings are in line with previous studies, such as those by Montgomery and Rodríguez Ledesma, and show that the NSR patterns differently in northern Middle English and Older Scots, the latter including areas having Gaelic as their first language. Some other conclusions can be drawn from the study; for instance, the fact that the oblique pronouns (us and them) behave in the same way as the subject pronouns with regard to the operation of the NSR. In all the cases where the object pronoun was used, the verb took the inflection -s, as they always were non-adjacent to the verb. Additionally, an interesting fact has been observed in the first and third person: the conjunction or seems to trigger the use of the inflection -s in cases where, according to the NSR, the verb should take zero inflection since the personal pronouns are adjacent to the verb. Although 41 there is no sufficient evidence of this (only 8 examples in the present study), it is an interesting fact that has not been observed in previous research. The study’s main strength is its thorough approach, considering all available examples and contexts of first, second and third person plural forms. Even the information that was not significant for the analysis was tackled in the analysis. This inclusiveness allows for a well-rounded analysis of how these forms function in various linguistic environments. In addition, the analysis carried out was quantitative in an attempt to avoid any possible bias. However, the paper faces several challenges that result in some limitations. For instance, due to the small number of examples of second person plural forms, this context is underrepresented compared to the abundance of first or third person plural examples. Although this is expected in the genre of the texts of the analysed corpus, as they are mainly legal and formal texts, this limitation affects the study’s ability to fully explore second person plural forms, especially when testing the Type of Subject Constraint. Finally, it is worth emphasising that this study fulfils its aim of contributing to the literature on the operation of the NSR in Older Scots. It successfully answers some previously asked questions, and it might be of help for further investigation. In addition, it also allows for the continued exploration of other issues, like the influence of the conjunction or in the use of the inflection, the origin of the NSR, and so on. 48 Appendix II. Second person plural syntactic contexts /vps22<P+ 0 15 11 /vps22<P- +is 1 1 /vps22<P->= +is 1 1 /vps22<P^n+ +IS 1 1 49 Appendix III. Third person plural syntactic contexts /vps23 +IS 3 2 /vps23<P&P+ 0 1 1 /vps23<P+ +is 1 1 0 85 69 /vps23<P+>= 0 2 2 /vps23<P- +IS 6 6 +I[S] 1 1 +is 2 2 0 1 1 /vps23<PO- +is 1 1 /vps23<P^P+ +IS 2 2 +is 3 3 /vps23<P^P- +IS 2 2 +I[S] 1 1 +YS 1 1 +is 2 2 /vps23<P^n+ +IS 3 3 +is 1 1 /vps23<P^n- +YS 1 1 /vps23<P^n^n- +is 1 1 /vps23<Pn^n- +YS 1 1 /vps23<Pn^n->= +YS 1 1 /vps23<R+ +ES 1 1 +IS 37 30 +ISs 2 2 +S 2 2 +S~ 1 1 +YS 8 7 +is 38 32 +z 1 1 0 1 1 /vps23<R+<= +JS 1 1 /vps23<R- +IS 32 26 +JS 1 1 +S 1 1 +YS 13 9 +is 6 5 /vps23<R-<= +YS 1 1 /vps23<Rcj+ +is 1 1 /vps23<cjR+ +IS 1 1 +is 3 3 /vps23<cjR- +IS 42 42 +S 2 2 +SE 1 1 +YS 6 6 +YSs 1 1 /vps23<n&P+ +IS 1 1 /vps23<n&n+ +IS 5 5 +YS 2 2 +is 13 12 /vps23<n&n- +IS 9 8 +S 1 1 +YS 4 4 +is 19 17 0 1 1 /vps23<n+ +IS 34 29 +JS 1 1 +S 1 1 +YS 5 5 +YSs 1 1 +is 31 22 /vps23<n+{sic} +YT 1 1 /vps23<n- +IS 52 35 +ISs 1 1 +S 1 1 +Ss 1 1 +YS 12 11 +YSS 1 1 +is 41 33 0 3 3 /vps23<n^n +is 1 1 /vps23<n^n+ +IS 4 4 +JS 1 1 +S 1 1 +YS 2 2 +is 1 1 /vps23<n^n+>= +is 1 1 /vps23<n^n- +ES 1 1 +IS 3 3 +S 1 1 +YS 1 1 +is 2 2 /vps23<n^n^n+ +YS 1 1 /vps23<n^n^n- +IS 2 1 +is 1 1 50 /vps23<nn&n+ +IS 1 1 +is 1 1 /vps23<nn&n- +IS 3 2 /vps23<nn^n+ +IS 4 2 +S 3 3 +YS 2 1 +[]S 1 1 +is 4 4 /vps23<nn^n- +is 2 2 /vps23<nnnn- +IS 2 1 +is 1 1 /vps23>n+ +IS 1 1 +is 1 1 /vps23>nnn&n+ +is 1 1