History of Scots to 1700

7. Grammar (CM) [114]

The value of DOST as a source of grammatical information should not be underestimated. Obviously, certain aspects of grammar do not lend themselves to an alphabetical treatment, but many do, and all of the function words are fully treated in DOST, as well as the grammatical behaviour of lexical items.

7.1 Northern Middle English

The grammar of OSc is substantially that of nME. It shares in the many developments that arose from the contact between OE and ON, some of them direct borrowings, such as the th- forms of the third person plural pronouns, others precipitated by the unstable linguistic situation, notably the loss of most of the OE inflections,[115] and perhaps the fronting of Vowels 4 and 7 (Samuels, 1985; see §§6.1, 6.10). It has been suggested that the close relationship between the two Germanic languages made them mutually intelligible at this time. The relationship was appreciated by a contemporary observer:

Englishmen write English with Latin letters such as represent the sound correctly. ... Following their example, since we are of one language, although the one may have changed greatly, or each of them to some extent ... I have framed an alphabet for us Icelanders ... (c1150, quoted in translation by Skeat, 1887: 455)

Much of the core vocabulary of everyday peasant life was the same, e.g. man, wif(e, folk, hous, under, mine and thin(e (or obviously cognate, as with kirk = church, etc., see §, and all the more so if learners disregarded the details of inflectional endings.

The ON influence is remarkable for its pervasiveness in the Scots and English linguistic systems. It is quite easy for languages to borrow items of vocabulary from each other, since vocabulary is relatively open-ended and atomistic. If borrowing is extensive, it may affect the phonological system (as with French influence, giving Vowels 9 and 10 and the palatal consonants). But it is much rarer for the morphological systems to be affected, because morphological structures are relatively closed and tight-knit. The influence of Latin on Scots and English grammar, for instance, is peripheral, and is much more marked in certain prose registers, where the vernacular replaced Latin, taking over functions in the process. ON influence on the grammar, in contrast, affects the language in its entirety. In addition to the personal pronouns and ordinal numerals discussed below, other grammatical words were borrowed, including er(e, a minor variant of are (s.v. be A II 7), thoch(t (replacing OE þeah), at 'that', fra 'from', and man 'must'. The -and inflection of the present participle is likewise from ON -andi.

Prior to the establishment of a prestigious, standardised, form of the vernacular based on London, innovations spread outwards from the dynamic nME dialect, in the early ME period. Eventually, many nME forms (including the ONhb -es ending of the present tense, but not the complex environmental constraints that governed its use in nME as in Scots) became part of StE. Consequently, as medievalists from Skeat onwards have observed, Barbour is more accessible to the modern reader than his contemporary, Chaucer:

... the Northern dialect [of Middle English] dispenses with inflectional suffixes more than either of the others [the Midland and Southern]. This it did at so early a period that poems in this dialect often present a curiously modern appearance, and would do so to a still greater extent if it were not for the frequent introduction of Scandinavian words, many of which are now obsolete in our modern literary language [i.e. StE]. In other words, the difference between the Northern English of the Middle period and the English of the present day lies rather in the vocabulary and in the pronunciation than in the grammar. Barbour's Bruce is as old as the poetry of Chaucer, but has a more modern appearance. (Skeat, 1887: 34)

In the MSc period (corresponding to EModE), Scots followed French and English in a process of elaboration ultimately modelled on Latin. As Scots and English became more similar to French and Latin, through the borrowing of syntactic structures and idiomatic constructions, Scots also drew closer to English, often lagging slightly behind the latter in the adoption of new structures, whether borrowed or innovated (Görlach, 2002).

7.2 Loss of inflections

OE inflections had a number of different vowels, which were later replaced by a single vowel, written &ly;e>, conventionally interpreted as /ə/. The /m/ of -um (for instance, in the dative plural of nouns) became /n/. Subsequently, final /n/ was lost from many inflections. These changes, amongst others, define the transition from OE to ME.

Final -e was lost in the ME period (subsequent to OSL, see §6.6), earliest in the north. Chaucer uses -e optionally to fit the metre, but Barbour does not (see further, § In the north[116] <e> changed to <i>, conventionally interpreted as /ɪ/, starting in the 13th century.

According to Aitken (2002: §13), final -e must have persisted in PreSc till near the end of the 13th century. But by the early 14th century it had been deleted, apparently in all environments. Aitken cites the evidence of early place-name forms from DOST:[117]

(1) Nouns with -e# as part of the stem:

Galtunesside 1143-7 (OE sīde), Kaldewelle c1190 (OE wella), Aldetuneburne c1200 (OE burna), [118] Bradestrothirburne c1220, Bradewude a1240 (OE wudu), Farenyacredene c1320 (OE dene); and numerous examples s.v. kirk;

(2) strong feminine nouns with inflectional -e generalised:

Holemede c1200 Melrose (OE mǣd fem., mǣdwe ‘meadow’), la blac Rode 1291 (14th century) (OE rōd fem.), Redehalle 1373-4 (OE hall fem.).

Contrast with both (1) and (2) the historically correct omission of final -e in:

Blakepol c1190 (OE pōl ), Wytefeld c1200 (OE feld), Kaldestrem c1200 (OE strēam), Fitheleres flat 1226-34 (ON flatr adj.);

(3) weak forms of adjectives (plurals are not available): weak forms of adjectives with monosyllabic stems are almost always spelled with <-e> in the DOST material before the 14th century, e.g. s.v. ald, blake, hare, quhite, red(e (see the examples cited above), and in <holleresky lech> l214 (OE hol adj.) (s.v. lech(e n.3). But after vowels, as in resky adj. ‘overgrown with coarse grass’ (s.v. reiskie adj.), the -e had been already lost, it seems.

He also presents the evidence for -e deletion:

(1) For nouns, spellings directly showing -e deletion include: blod(e)wyt, bludwyt 14th century (OE wīte) (s.v. bludewite n.), kyrkhalch (s.v. kirk n.).

(2) Uninflected attributive adjectives become common from the late 13th century: Westfeld 1294, Hwytfeld 1333, Caldclogh 1363, red heuch 1388-9, Redhowch c1220 (14th century).

The 1317 Aberdeen Court Roll[119] shows stem -e deletion in the surname <wytleyir> ‘fault-finder’ (OE wīte) and the place-names <kyrcgat> (OE cyrice, ON kyrkja; ON gata) and <gallugat> ‘gallows-road’. In the place-name <grendoun> the element <gren> represents OE grēne ‘green’. By the Scone Glosses of c1360,[120] the incidence of <-e> is not predictable on etymological grounds.

7.3 Personal pronouns


The grammatical concepts expressed in the personal pronoun are number, case and gender. Gender was originally on a rather arbitrary basis. The OE definite article agreed with the grammatical gender of the noun, although pronouns were usually in the natural gender. Gender conformed in the early ME period to natural gender, so that inanimate objects are now treated as neuter, and masculine and feminine literally refer to male and female. This was apparently brought about by the loss of inflections, and took place first in the north (Mustanoja, 1960: 43-8).


The OE personal pronouns were as follows (ignoring the dual number, which was found in first and second persons):


The native forms of the third person plural - hī, hira, him - were replaced within the Great Scandinavian Belt (see §2.2.3) by ON forms with th- (also written <Þ>) – thai, thaim, thair, a replacement that later spread, of course, to PreStE as well as to Scots. The nominative form, thai, was the earliest to be adopted in nME, possibly helped by the fact that native OE Þā (a part of the definite article, used as a demonstrative) already overlapped in some senses with (OED, s.v. they). In sME, however, Þā develops regularly to tho, later altered to those, quite distinct from they. In Scots, it appears that the reflex of Þā converged with thai, the personal pronoun (s.v. tha(i), with both sharing Vowel 4 and Vowel 8 doublets (Aitken, 2002: §22.4).

The origin of the third singular feminine /ʃ/ forms (scho, she, etc.) is still debated. One possibility is that hēo underwent a dialectal ON sound-change (Dieth, 1955, but for the arguments in favour of a native development, see Britton, 1991).[121]

The nME ik form of 'I' is found in Scots uniquely in the Edinburgh MS of The Bruce. This either escaped palatalisation (in contrast to sME ich) or took the consonant of ON ek.


The OSc personal pronouns were as follows:


7.3.4 Possessives

Perhaps because the idea of possession does not apply in the same way to inanimates, the form its does not occur.[122] The periphrastic possessive of it (cf. below) is, however, found, although the most common expressions are thareof and of the samin/same (s.v. it pron. 2 b). The uninflected possessive it, recorded for ModSc, is only rarely attested (s.v. it pron. 3).[123]

My is a reduced form of mine, used attributively (i.e. pre-modifying nouns). In OSc, mine (q.v.) was also still used attributively, but after the 15th century, attributive mine occurs only before vowels and /h/. Correspondingly, my (q.v.) is at first more common before consonants, but does also occur before vowels from an early date. Myn an ‘my own’ gives nane (q.v.) by wrong division. Thi arises similarly from thine. Thine occurs only before vowels and /h/ in the DOST corpus (the data for the second person, especially the singular, are of course limited in the written corpus of the language).

Predicative forms with -s - hirris, ouris, ȝowris, thairis - like so many of the developments of this period, appeared first in early nME.

7.3.5 Possessives post-modifying nouns

As in EModE (but perhaps not so commonly), his could be used instead of the possessive inflection, e.g.:

the Kingis hienes his commissionaris (s.v. he pers. pron. 3 c)

and likewise the analogous thair(e (q.v. 2), but not apparently hir,[124] although the construction goes back to OE with his and hir.

7.3.6 Second person pronouns

In the second person plural, ȝe is the nominative form and ȝow the accusative until they begin to fall together in late MSc. This takes place considerably later in Scots than in English (where you had replaced ye by the end of the 16th century). The earliest examples of nominative ȝow in Meurman-Solin's corpus of letters[125] are from the Crim. Trials (as early as 1567), but it does not overtake ȝe until the 1640s in women's letters, later in men's letters (2001).

The old singular forms were abandoned in 16th century London, and subsequently in most English and Scots dialects, but are still retained as familiar singulars in the traditional dialects of parts of the north of England and of Orkney and Shetland. As in many modern European languages, the second person plural was used in OSc as a polite, deferential, distant form of address in the singular. The second person singular forms (thow, etc.) were used as a familiar, warm or patronising form of address (cf. thow pron. 3), and were normal from a senior person to a junior or subservient one.[126] It is probably indicative of coldness, therefore, when the father in the 1551 quotation (s.v. ȝe pron. 1 a) gives his son ȝe.

However, in Biblical translation, thou and ȝe were used literally to translate the singular and plural of Greek and Hebrew, and this simple singular v. plural usage is found in religious contexts in OSc, though not always consistently:

O sueit Lord, I desyre richt hertly to thanke ȝou for Þe gret pane Þat ȝe had in Þi precius body. (Ane Dewoit Exercicioun ll. 283-4, from Bennett ed., 1955)

As in ModSc, the is also found as the nominative form, but there is no evidence of the modern Orcadian use of thoo as the accusative.

7.3.7 Third person plural used for singular

As in modern colloquial usage, thair(e is used to recapitulate noun phrases containing ilk, every and na, where the reference is semantically plural; and there are also occasional instances where gender cannot be specified:

A man or woman being lang absent fra thair party (s.v. thair(e possess. adj. 1 e).

7.3.8 Indefinite personal pronouns

From the late 16th century on, there are examples of ȝe used as an indefinite personal pronoun (ȝe pron. 4). Ane 'one' in this use is apparently not idiomatic to OSc (any more than to ModSc or ScStE).[127]

7.3.9 Impersonal constructions with the dative

The main use of the dative, when it existed as a separate case-marked form in pronouns and nouns, was as the indirect object. It was also used to indicate the person concerned in relation to impersonal verbs. Since the dative and accusative had fallen together as a single oblique case, it is this rather than a nominative pronoun that precedes an impersonal verb in OSc, e.g. you likis, rather than ye like, and so on. Verbs used in this way in OSc include ((be)fall, behufe (but by MSc mainly with the nominative), forthink ‘regret’, happin, like v.1, list v.1, nede, think ‘seem’, worth ‘to have need’ (s.v. worth v. 3).

7.3.10 Pleonastic ‘it’

Pleonastic (redundant) it is used in various constructions, including impersonal ones (cf. ModStE ‘it rains’). It is also used where ModStE would have there as a dummy subject:

It was na neid to bid him strike (early 15th century, s.v. it pron. 4 b).

(This continues in ballad Scots and occasionally in modern folk narratives.)

7.3.11 Reflexive pronouns

In OE, reflexivity was expressed by a personal pronoun in the dative, or by dative or accusative + the corresponding form of the adjective self. Both alternatives continue in OSc (i.e. an oblique pronoun with or without following -self). Selvin, the form with the originally dative ending, varies with -self in ESc (as in ME down to the end of the 15th century - Mustanoja, 1960: 147), and as an archaism in Douglas (s.v. himselvin pron., hirself pron. (a)).

As in sME, new forms appear with genitive pronouns, on the analogy of herself (where the original dative and genitive happened to coincide in form), the adjective self being reinterpreted as a noun (cf. §7.6.4). The original me-self (q.v.) is occasionally written in more colloquial OSc texts, but the usual written form is my-self.[128] Hissel had apparently not yet developed in OSc to replace himself. In keeping with the absence of its (above), there is no its self. Itself (q.v.) is also rare, the usual expression being the self (s.v. self C 2).

The plural pronouns were originally followed by the appropriately inflected form of the adjective. The dative plural survives only in ESc as thaim selvin (s.v. thaim self pron.), alongside thaim self and thair self; and ȝour selwyn in Barbour (s.v. ȝowrself pron. 1 a). The less-well documented our-self shows no -selvin forms. In sME the -selvin forms were reinterpreted as noun plurals, and new -es forms were created. The corresponding -is forms in Scots - thaim selfis, thair selfis, our-selfis, ȝourselfis - are mostly found after about 1540. Uninflected forms also continue as an alternative. (See also §

7.4 Nouns


In OSc, the grammatical categories expressed in the noun are number and case. Surnames can be plural, e.g. Jhon and Andro Moffatis (Aitken, 1971: n.12). For adjectives in concord with plural nouns, see §7.7.3. Some specific adjectives mark the difference between mass and countable nouns (see §7.7.4). Interrogative and quh- relative pronouns (the latter not at first) distinguish between human and non-human nouns as antecedents (see §§7.9,


OE had a number of noun declensions, but in OSc, as in late ME, virtually all nouns have been transferred to a much simplified descendant of the OE general masculine declension, thus:


7.4.3 Ø plurals

OE had the Ø (zero) inflection in the nominative plural in some general neuter nouns and some minor declensions. This survives in a few words, e.g. thing (cf. ModSc awthing, etc.), folk (alongside folkis), and some animal names, such as s(c)hep(e, dere n.1, hors (joined by horssis from the early 16th century), and foul and fisch in the collective sense. The variably zero plural of vers(e is carried over from French.[129]

7.4.4 Mutated plurals

In a small number of words, different vowels are found in singular and plural, e.g. fut(e/fete (but see §7.4.8), mous/mise.[130] Again, there is Ø inflection in the plural.

7.4.5 -n plurals

A few OE -an plurals survive to give OSc plurals in -(i)n, e.g. oxin or owsyn (plural of ox), ene (plural of e).

S(c)ho n. is transferred to this declension from the 13th century on, thus OSc schoyne alongside schoys.

Brother had an uninflected plural in OE. The plural brether is from the dative singular.[131] From MSc on, the bretherin type -n plural is also found.

7.4.6 -r plurals

A few OE -ru plurals survive, e.g. childer (s.v. child n.), cair (q.v., plural of calf, with loss of /v/). OSc also has the double plural childrin. The plural lamber of lam (q.v.) occurs in place-names.

7.4.7 Ø genitives

A minor declension in OE (of nouns of relationship ending in -r) is uninflected in the genitive singular, and there are occasional survivals in OSc, e.g. sister son ‘nephew on the sister's side’, brother dochter (q.v.) ‘niece on the brother's side’.[132] See also fader n. 1 b, lady n. 6 b, maister, moder n.1 2. Zero genitives are also found variably with personal names (Aitken, 1971: n.3, citing Müller, 1908: 121), and there are examples with titles in sentences quoted by Moessner (1997):

the kyng of ingland saue conduct (p.119)
to the erl of Herfurd cosyne (p.121).

7.4.8 Expressions of quantity

In certain expressions of quantity and extent, an uninflected plural noun is found (often varying with the inflected plural), which descends from the old genitive plural (used as a partitive genitive). The inflection was -a in most declensions, giving Ø with the loss of inflections. Some examples are:

For ii pund of butter (s.v. pund n.1)
Off haylstanys than ane fell ... awcht fute brayd (s.v. fut(e n. 4 b (2)).

See also e.g. eln(e, lade n.1, mark n.2, mile n.1, ȝer(e. This usage is extended to other nouns of measurement, e.g. dosane, gros n.2, scor(e. A plural form of fute as a noun of measurement is also used: it is commonly futis (s.v. fut(e n. 4 b - d).

More commonly than in ModStE, expressions of quantity have two nouns in direct apposition (cf. ModStE ‘a few berries’ as opposed to ‘a handful of berries’):

For vj paris schone (s.v. pair n. 1 a sing. (1))
Welcum confort of alkynd fruyt and grayn (s.v. alkind adj.).

But also e.g.:

For xl pare of schone (s.v. pair n. 1 a plur. (2)).

Alkin(d) apparently occurs with following of only in the construction alkyn kynd of:

all-kyn kynd off Inglis men (s.v. alkin adj. 1 c).

It has been suggested that the decline in the use of apposition was due to Romance influence (see Sørensen, 1957: 147).

7.4.9 Periphrastic datives

With the loss of distinct dative forms in early ME, periphrastic means of expressing the indirect object are increasingly used instead, mainly to and for plus noun or oblique pronoun (as in ModStE ‘give the book to Mary’ as well as ‘give Mary the book’).

7.4.10 Periphrastic genitives

Periphrastic forms also become common alongside the inflected genitive. French and Medieval Latin may have been influential in this respect, as periphrasis with de is the normal form of the genitive in the languages descended from Latin. The periphrastic genitive becomes the most common type in Scots by the 14th century. The fact that it is more strongly favoured in prose than in poetry in the early stages is suggestive of Romance influence (see Mustanoja, 1960: 74-8).

7.4.11 Split genitive

When a noun phrase to be put into the genitive itself contains another of phrase, split genitives are an alternative, at least with reference to titles and family relationships, as in:

The Kingis dochter of Nuby (s.v. king n. B 3 a (3))
Emma his latter wyfe .. Ducke Richardis dochter wes of Normondy (s.v. duke n.1 1 b).

7.4.12 Adverbial dative

By the OE period, the dative had almost entirely absorbed an older instrumental and locative case. Quhilum and seldin are such datives surviving as adverbs (see Mustanoja, 1960: 104).

7.5 Articles and demonstratives


The indefinite article a, an developed during the OE period from the numeral ān ‘one’. From the 12th century on, an was weakened to a except before vowels. In MSc, ane is written before consonants as well as vowels, but there is no evidence (for instance, from modern dialect speech) that the /n/ was pronounced in this position. This is a trap that anyone reading OSc aloud should beware of.


The definite article the (already a nominative masculine singular form of 'the' in ONhb) replaces the fully inflected definite article se of OE.

The Old English definite article was declined as follows:

nominativese, ÞesēoÞætÞā
instrumentalÞȳ Þȳ 

The demonstrative that (see below) descends from the nominative neuter form. Þæt is reinterpreted, by misdivision, in the ta 'the one' and the tother ‘the other’ (see ta, tothir), but not normally in that ane (see that B 1 d).

Similarly, Þan (the later form of OE Þǣm), is wrongly divided in for the nanis (see nanis).

For-thy ‘because, therefore’ retains the instrumental form of se (cf. §7.4.12).[133]


OSc shares with English a number of usages that survive in modern Scots and English dialects, but not in ModStE, or not with such a wide range of nouns: see the def. art. I 2, 3 e, 8, 9, 10, 11, and cf. OED the dem. adj. 3 b, 10 c, 2 c, 5, 3 d, 8, 7. The morn 'tomorrow' and the day 'today' are attested first in nME (OED s.vv. morn 3 c, d, the 2 c); other combinations develop later in Scots and nME. OED attaches no label to its 2 b, but all the examples of the before cardinal numerals denoting years are from Scottish sources (see DOST the 3 e).

7.5.4 Demonstratives

That was part of se 'the' in OE (see above). This comes from a different OE word, Þes, again fully inflected (Þis being the nominative neuter singular). The use of that and this followed by a plural noun, e.g. that wayis, now characteristic of Northern and Insular Scots, was not so geographically restricted in the past (see that B 1 b, this A 1 e, A 2, A 3; and SND s.vv. dat, dis, that, this; MED s.v. that adj.[134]).

The use of that as a demonstrative adverb, meaning 'so', e.g. that mekyll, appears only at the end of the 16th century. It is first attested (mid-15th century) in PreStE (OED s.v. that dem. pron., adj. and adv. III), although examples after the 17th century are colloquial, and it is now considered non-standard (and is very widespread in English dialects as well as being the usual ModSc form: see EDD s.v. that 11). This is the sociolinguistic profile of a feature that diffused to Scots as a spoken form, as opposed to a literary anglicism.

Tha ‘those’ descends directly from the OE plural form Þā, but converges, as we saw, with thai 'they'. There is a rare OSc form thais (presumably by analogy with English those), mainly in the 17th century.

The origin of the plural thir ‘these’ is obscure. Although ON þeir is most likely, and the word has a northerly distribution from the outset, there are difficulties with this derivation: the ON word means 'those', not 'these' (as well as meaning 'they'), and is borrowed (as thai) in the sense 'those' (and as 'they'); the -r is an inflectional ending, which would almost always be dropped when borrowed (see further OED, MED s.v. thir).

ȝon(e, indicating something more distant than that, from an OE adjective eon, is also found as a demonstrative in Scots (as in ME).

7.6 Numerals


There are two parallel series of numerals: cardinal and ordinal. The main forms are given below:

Cardinal numeralsOrdinal numerals
OScOscOE (ONhb)[135]ON[136]
a, ane firstforma, fyrmestfyrstr
twa, twane (as adj., only verse)secundōþer, æfterraannarr
threthrid (q.v.), thredðirdaþriði
fourferd(e, fourtfēarðafjórði
fivefift fīftafimti
sex, saxsext (q.v.), saxtsiexta, sestasétti
sevinsevynd, sevint (q.v.), seyntseofunda, siofundasiaundi
aucht auchtad, aucht, auchtandeahtoþaátti
nyn(enynd(e, nynt(enigoþa, nigeþaníundi
ten te(i)nd (q.v.), tenttēoþa tíundi
twelf (q.v.), twalff, twel(l, twal(l, twoll twelft(e (q.v.),twal(l)t, twolttwelftatólfti
thretten(e thretteind, thrette(i)nt (q.v.)þrēotegeðaþrettándi
fourtenefourtent fjórtándi
fiften(e fiftende, fiftent(e fimtándi
sexten(e (q.v.), saxten(esextend(e (q.v.), sextent, saxteintetc.
sevinten(e sevintend (q.v.), sevinte(i)ntsiaut(i)ándi
nyn(e)ten(enyn(e)tende, nyn(e)teintnítiándi
twenty, twinte, twantie, twontytwentyd, twentiand (s.v. twentieth) twēntigoþa tuttugandi, -undi
thret(t)ye, thrity (s.v. thret(t)y)thret(t)yd, threttiand (s.v. threttieth)
sexty (q.v.), saxty, threscore- (sextieth)
auchty, fourscore-
nyn(e)ty -
hundir, hunderd, hundert, hundreth (= 100 or 120 according to context) hundreth(e
thousand(e (q.v.), thowsant, rarely mill(e-

Forms with <o> s.vv. twelf and twelft(e can be compared with their MLG cognates, but the development to /o/ Vowel 18 is basically a native one, from /a/ Vowel 17 between a labial and /l/, cf. e.g. follow n. (see Aitken, 2002: §14.18).

The suffix of the ordinal was varied both in OE and ON. There has been a process of levelling, leading to the generalisation of the -t ending (as in OE siexta, etc.) on the ordinals between 4th and 19th (cf. the generalisation of -th in English), thus nynt(e as well as nynd(e in ESc and later fourt as well as ferd(e . Ellevint, remodelled on the cardinal numeral ellevin, replaces nME elleft (< ONhb ællefta), of which there is apparently no trace in OSc (see OED s.v. eleventh). Nor do we find any forms other than -e(i)n + -d (see below) or -t for the ordinals between 13th and 19th. Aucht ‘eighth’ goes back to ON *ahta, and replaces auchtad. Forms in -th are anglicisms, as apparently are eicht (see §4.2.1), and teith (s.v. te(i)nd).

The OE -þ- forms have given /d/ rather than /θ/ or /đ/[137] in ferd(e and in -tyd (< -tig(o)þa): see fourtyde, thret(t)yd (s.v. threttieth) and twentyd (s.v. twentieth).

ON forms appear to lie behind nynd(e and te(i)nd and -ende as in fiftende, nyn(e)tende, sextend(e, thretteind (s.v. thrette(i)nt). The ON ending survives in the variants auchtand (q.v.), threttiand (s.v. threttieth) and twentiand (s.v. twentieth).

The extent of ON influence in the ordinals is less apparent than in the pronouns as few of these forms survive, largely because of the later generalisation of -t.

As cardinal numerals, with reference to dice, we also find (from OF) ace, dewis (recorded only in the collocation dewis ace), tray n.2, sink n.2, and sice. It is probably safe to assume that quatre was also used, as in contemporary English (see OED s.v.).

As nouns of enumeration, dosane, gros n.2 and scor(e (with or without of) were also, of course, in regular use, the first two in particular mainly for tradeable commodities, but also, in the case of dosane, for strength of manpower (also a common use of score). The use of the score is also reflected in the long hundred of six score[138] (see 1687 quotation s.v. scor(e n. 1 c (2), and also hundir num. 2 a, hundreth num. 5 and gros hunderth, and cf. short hunder s.v. hundir num. 2 b). In certain places, the mais(e (q.v. n.1) was used for fish.


The cardinal numbers from thre onwards are quite commonly used as ordinals,[139] e.g.

The fyv day of ... the yer of our Lord ... nynti and twa (s.v. five num. 3)
The nyne part of ane nettis fischin (s.v. nyne num. 2).

Twa is used in this way only in the combination twenty twa (s.v. twa num. 5). This usage presumably originates with the literal expansion of similar expressions written with figures. The most frequent contexts are those that are most frequent generally for ordinals: measurements of time (see examples with figures s.v. day n.1 4 b), and fractions.[140]

As usual when sound and symbol do not match, there are reverse spellings, with ordinal endings on cardinal numbers: s.v. ellevint num. adj. 2, nyn(e)tinth, sevintend adj. b, sex (for sext, saxt variants), sextieth, thretten(e (for -inth variant). The fact that the ordinal ending could be written as a superscript <t> may have contributed to these interchanges (cf. §3.3.2).


In composite numbers between 21 and 99, the usual form is apparently ‘tens and units’ (not, as in contemporary ME, ‘units and tens’, though this is also found):

The ȝeir of God ane thowsand fyf hundrycht fyfty and nyne ȝeiris (s.v. five num. 1 (b)).

It has been suggested that this is due to Romance influence (Sørensen, 1957: 148).

7.6.4 Ane

In certain contexts, ane is used in a vaguely emphatic way, thus on-ane (q.v.) in rhyme (also anglicised anone, onone forms), and mony ane (s.v. mony).

Allane (all + ane) is a strengthened form of ane. In expressions such as him alane (occasionally also him ane), it reinforces an oblique (earlier dative) form used in a reflexive sense (s.v. allane, ane numeral and adj. A 3; Mustanoja, 1960: 100). Already in OSc, this could be replaced by a genitive, his allane, and with the reduced form lane, his lane, etc. (s.v. lane adj.; Mustanoja, 1960: 293ff.; and cf. §7.3.11).

The superlative of ane (also nME) is thought to be a calque on ON einna ‘of all’, literally ‘of ones’ (s.v. ane numeral and adj. A 5 c).

7.7 Adjectives


In OE, adjectives were declined for number, case and gender, in agreement with the following noun. There were, furthermore, two patterns of inflection, according to whether or not the sense was definite. Adjectives are definite (or weak) mainly when used attributively together with a demonstrative or the definite article (e.g. ‘this green cloth’).

Indefinite (strong) forms were uninflected in the nominative singular, masculine and feminine. Definite forms, however, were inflected thus in the nominative singular:


This gives a final –e in early ME. In PreSc of the 12th and 13th centuries, place-names sometimes show this final –e (see §7.2).

A fossilised remnant of the genitive plural remains in alther (q.v.).

7.7.2 Comparison

There are two methods of forming comparatives and superlatives of adjectives: the inflections –ir, -ist, and the periphrastic mar(e (or ma) and mast(e. It has been suggested that the periphrastic type is modelled on French plus, le plus and/or Latin magis, maxime. Against this, both methods are applied freely to both native and Romance words. Nor is the periphrastic type particularly associated with prose (see Mustanoja, 1960: 278ff.) At the least, we can say that Scots (with English) again finds support from French and Medieval Latin in its development towards more analytical structures.

Comparison of two equivalent things was expressed in OE by swā ... swā ‘so ... so’. The first swa was sometimes strengthened by all ‘all’.[141] ONhb allswā gives OSc alswa, alsa, als and as. The usual construction in OSc is als ... as, less commonly as … as, sa … as (s.v. als adv. and conj. 2, as adv. and conj. 5, sa).

In other types of comparative clause, the conjunction of comparison is variously as, e.g.:

My truble ... is hauiar as the sand of the sea (s.v. as adv. and conj. 6)

or or conj.1, prep.1 3, na conj.3 , nor conj.2, then.

7.7.3 Inflected plurals

Inflected plural forms of adjectives are sometimes found, under French and Latin influence, particularly in scientific and legal texts, where the plural form is often preserved in particular fixed collocations. This construction is less regular and more idiosyncratic in literary prose, "even … under the immediate influence of a French or Latin original" (Ledesma, 1998). With Romance adjectives, the adjective is usually placed after the noun (following Romance word order), e.g.:

of many wrongs subtiles and also open oppressions (quoted by Ledesma, 1998: 25).

The only native adjectives that are commonly inflected are quhilk (see §, other, e.g.:

Send ane boy ... witht ane letter of manteinance and otherris letterris (s.v. other pron.1 A 2 d (a)),

luvit (s.v. also belovit) and for(e)said, said. There are also rare plurals, thairz and yhourz, of the possessive adjectives thair(e (q.v.) and ȝowr (q.v.). (Fore)saidis is influenced by plurals of French le dit and Latin (prae)dictis (Mustanoja, 1960: 277). Ledesma, following Sheppard (1936), suggests that otheris may be influenced by the corresponding pronouns: in some contexts, it would be difficult to distinguish between adjective + noun, and pronoun + noun in apposition, e.g.:

þe capitanis of þe tribis and vþeris þe wourthiest personis for þat tyme

and Ledesma adds that quhilkis is a similar case (1998: 27).


Mar(e and eneuch both vary in form according to whether they modify a mass noun or a countable noun. Expressing quantity of a mass noun, we have mar(e and eneuch:

A mar sowme of ane hundredth merkis ekit apon a new reuersioun maid (s.v. mar(e A adj. 1 b (1))
Wallace commaunde a burges for to get Fyne cawk eneuch (s.v. eneuch 1 adj.).

Expressing number of a countable noun, we have ma and enew:

No inhabitant ... shall ... invite anie ma persones to be gossopes [sc. fellow god-parents] ... bot four gossopes (s.v. ma A adj. 1 (1))
With hym ma men than enew (s.v. enew).

Ma and mare are from different OE sources. Enew is from an inflected form of the same word that gives eneuch (the inflection having provided an intervocalic environment, see §6.10.3).

7.8 Verbs

7.8.1 Present tense inflections

OE verbs were inflected in the preterite as well as the present tense, but be is the only verb that continues to preserve any trace of this (cf. ModStE ‘he was’, ‘they were’).

King (1997: 176) reconstructs as follows the present tense inflections of late ONhb:


ONhb -as/-es is the source of the generalised -is ending of OSc (see below). This northern ending likewise spreads in the course of the ME period to other English dialects, and is the source of ModStE -(e)s.

Southern ME had -eth in the plural and third person singular up to the 17th century (cf. the Authorised Version of the Bible). Midland ME took the plural ending -en from the OE subjunctive mood. These other endings appear occasionally in OSc in mixed dialects such as that of James I, and under Chaucerian influence (see §9.3.1). Meurman-Solin also finds that the -eth ending is adopted as part of the process of anglicisation, increasing in Scots while it is decreasing in English (1993a: 251).

7.8.2 Present tense concord

In OSc (and to some extent still in ModSc) there are two systems of concord between subject and verb in the present tense. In the terminology of Montgomery (1994), if the subject is a personal pronoun (the Type of Subject Constraint), and comes immediately before (or after) the verb (the Proximity to Subject Constraint), the inflections are as follows:


e.g. I keip, thow keipis, he/scho/it keipis, thai keip. This is similar to ModStE, but note that the second person singular takes the same inflection as the third person singular (but see be and have, below). Keipst, etc. occur only as anglicisms.

Otherwise, the inflected form is used with all persons and numbers:


If a personal pronoun governs two conjoined verbs, the first is affected by the contiguity of the pronoun, the second not:

it obscuris and diminucis
I renunce ... and takis
we teche ... and commandis (examples from Kuipers, 1964)
Þai sla our folk but enchesoune, And haldis Þis land agayne resoune (Barbour, Bruce, I: 487-8).

King (1997: 176-7) reconstructs the late ONhb paradigm (see above) to show that the first person singular would regularly yield Ø, and all that has to be explained is Ø in the plural;[142] and she explains the Northern Present Tense Rule (in Montgomery's 1994 terminology) as an accommodation between this paradigm and the tendency to generalise the ending throughout (as in some English dialects).[143] She suggests that since the generalising tendency obscures the category of number, it is perhaps more tolerable when the subject is a full noun phrase. Klemola (2000), however, expands on a suggestion of Hamp (1975-6: 73, addendum), who pointed out the similarity between the dual system of concord and the dual system in p-Celtic languages, whereby the third person plural takes the same ending as the third person singular (cf. the -is ending here) except when the subject is a pronoun. The dual system would thus be explained as a p-Celtic substratum influence, extended to the other persons that share the Ø ending with the third person plural. (The Ø ending still requires separate explanation.)

In the historic present tense, which is exemplified in a few colloquial texts, the inflection is used throughout the paradigm regardless of the subject and its position:

swa we continewes in drinking qhill euerie man was his pynt about (John Campbell's Complaint, ?1613, quoted by Aitken, 1978: 98).

7.8.3 Subjunctive

The preceding applies to the indicative mood. OE also had a separately inflected subjunctive mood for hypothetical and conditional statements. The present subjunctive was formed thus:


A distinct subjunctive is sometimes marked in OSc by being uninflected throughout (the normal outcome of -e(n), see §7.2):

‘Now ga we furth Þan,’ said Þe king, ‘Quhar he Þat maid off nocht all thing,
Lede ws and saiff ws for his mycht, And help ws for till hald our rycht.’ (Barbour, Bruce, VIII: 261-4)

Moessner (1997) gives examples of conditional clauses signalled by the subjunctive without a conjunction, e.g.:

Chaip he away, we ar eschamit (p.140).

7.8.4 Imperative

The imperative may be inflected in the plural:

Kepys ȝow fra disparyng (Barbour, Bruce, III: 200)


Wyrk ye Þen apon swylk wys, Þat ȝour honour be sawyt ay (ibid. II: 340-1).

7.8.5 Interrogative

The interrogative is inflected as for the indicative, but the order of subject and verb is reversed. (For the development of the do periphrasis, see below.)

7.8.6 The past tense and past participle inflection of weak verbs

Main verbs fall into two groups: 'weak' and 'strong'. The weak verbs form the preterite (the past tense form) and past participle by adding a suffix, in OE -ed; in ESc -id/-yd; later -it/-yt or, after vowels, -d. There is occasional confusion with the ending -at(e) from Romance sources, e.g. imbarcat(t) as a past tense or past participle of imbark ((Meurman-Solin, 1993a: 233).

The ending is often omitted in borrowed Latin past participles, e.g. educat, imput (v.1).[144] Omission of the ending is often found in fixed phrases (e.g. "it is statute and ordainit"), and is at first more frequent in formal texts, "especially in statutory texts close to the central administration" (Meurman-Solin, 1997b: 117). The omission of the ending spreads, at a low level of frequency, to other past participles with stem-final /t/. It is more likely when the following word begins with a consonant (Romaine, 1984; Meurman-Solin, 1997b).

7.8.7 Strong verb classes

The strong verbs change the vowel in different parts of the verb. In OE, the present, preterite and past participle had different vowels, as well as being distinguished by inflectional endings. Additionally, there were two different vowels in different parts of the preterite (first and third person singular taking one vowel, plural and second person singular taking another). In OSc, as in ME, the four-vowel system is reduced to three, or even two (the past participle still being distinguished from the preterite in many cases by the -in ending). The OE infinitive usually has the same vowel as the present tense, but if not, the infinitive vowel is later generalised to the present.

Some weak verbs also have different vowels in present and preterite/past participle through the accidents of sound-change, e.g. beseke, besocht, by i-mutation.

Strong verbs in OSc come down from OE, with a few, notably gif and get, from ON. They are much simplified in comparison with OE, as well as being affected by sound-change. The past tense now has the same vowel throughout, usually that of the first and third persons singular. The Germanic strong verbs are traditionally divided into seven classes according to the vowels (ablaut series) that they take. After the OE period, verbs sometimes acquire unexpected forms by analogy with classes other than the one to which they originally belonged. Some examples of the regular developments are given below. Preterite plurals are given when different from the first and third persons singular.

Class I

 InfinitivePret. sing.Pret. pl.Past Participle
OScdrivedrafe, drave*drivdrivin

Other verbs in this class include (a)bide, bite, flyt(e, glide, ride, rif(e ‘tear apart’, ris(e, schrive ‘make/hear confession’, slid(e, smite, strik(e, writ(e.

Class II

 InfinitivePret. sing.Pret. pl.Past Participle
OSccrepe*crepe (crape as Class 1, weak crepit)*crup‘creep, cropyn (cruppin)

Other verbs in this class include bede ‘pray, command’, begrete ‘weep’, brew, chese 'choose', cleve v.1, fle v.1 ‘fly’, flete ‘float’ (flote v.2, a weak verb, is also found), frese, grete v.1 'weep', lese ‘lose’ (cf. chese and chose), s(c)hute, ȝet(t ‘pour’.

Class III

 InfinitivePret. sing.Pret. pl.Past Participle
OScclimeclam (weak climmed)*clumbclumbyn (clum)

Other verbs in this class include begin, bind, carve, drink, fecht, find, grind, help, ring, s(c)hrink, sing, sink, spring, swel(l, swim, thring(e ‘crowd’, win, worth ‘befall’, ȝeld(e.

Class IV

 InfinitivePret. sing.Pret. pl.Past Participle
OScbair (bere < OE beran)barebereborn (bore)

Other verbs in this class include brek, scher(e, stele, tere and irregular cum.

Class V

 InfinitivePret. sing.Pret. pl.Past Participle
OScbidbad, bade*bede*bedin (biddin)

Other verbs in this class include ete, ly, se, sit, spek(e, tred (s.v. trad(e), weve, wrek(e ‘avenge’.

Class VI

 InfinitivePret. sing.Pret. pl.Past Participle
OScdrawdreuch (drew)*drowdrawin

Other verbs in this class include fare ‘go’, fla ‘flay’, gnaw, s(c)hake, s(c)hape, schave, sla, stand, swer(e, ta(k, (a)walk ‘wake’, wes(c)h(e.

Class VII

 InfinitivePret. sing.Pret. pl.
OScblawblewblawin (blaw)

Other verbs in this class include behald, craw, dele, drede, fall, grow v.1, hald, knaw, lede, rede, wax ‘grow’.

As in English, the number of strong verbs decreases after the OE period. A few weak verbs do acquire strong forms by analogy, e.g. knit, preterite knat, past participle knut or knittin; and cast, preterite cust, past participle castin. Ring meaning ‘reign’, from Old French, has been conflated with native ring ‘ring’, and has preterite rang, past participle rong(yn). However, the trend is overwhelmingly in the opposite direction.

As King (1997: 178) points out, the weak form gied (past tense and past participle of gie) is attested only from the 18th century, although it shows MSc loss of /v/ (see §8.2.2).

7.8.8 be and have

Like other verbs that take inflections, the verbs be and have (whether as auxiliary or main verbs) show a dual system of concord, determined by the nature and position of the subject (see above). Montgomery (1994) confirms the impression of Murray (1873) that the typically Northern system of concord (with the form of the 3rd person singular generalised under the conditions described above) spread to the verb be by analogy, and is found only variably in OSc. It is already found in ESc (late 14th century), but does not reach 50% in Montgomery's corpus[145] before it is over-taken by the process of anglicisation. Montgomery's data for the operation of the Proximity to Subject Constraint are too limited for quantification, but confirm that this could also apply, e.g.:

I haif bene merwellus ewill wexit with my infyrmetie this v dayis passit and is nocht weill as yeit (p.90).

Be also preserves two past tense forms, wes and were,[146] again used according to the dual concord system, again already in late 14th century texts, again variably (Montgomery, 1994: 91-2).[147]

Unlike other verbs, be and have have distinct second person singular forms, art and haist respectively. However, is and has also occur for the second person singular (? in more colloquial texts), and it is these forms that survive into ModSc (see have, be, thow; SND hae, is, thou). Is as a second person singular form goes back to ONhb (SND s.v. is).

Adjacent pronounOther
singular 1I havehavis
singular 2thow haist, havishavis
singular 3he, scho, hit havishavis
pluralthai havehavis

Reduced forms of havis – has, etc. – are also found.

Adjacent pronounOtherAdjacent pronounOther
singular 1I amisI weswes
singular 2thow art, is [148]is, beisthow weswes
singular 3he, scho, hit isis, beishe, etc. weswes
pluralthai aris, beisthai werwes

The present subjunctive is be.

7.8.9 Infinitive

The infinitive expresses the bare sense of the verb. In most contexts it is introduced by to or for to:

To se that selie mous it wes grit sin (Henryson Fables 299)
Al Cristine man fore to sla (s.v. for prep. 11 (2)).

To also has the form til(l from ON. (This enters StE only as till ‘until’.) In ESc, to occurs only before consonants, with til before vowels and /h/.

The ON preposition with the infinitive was at. This is found in Northern English, but in Scots (as in StE) only in fossilised phrases, particularly ado (at + do ‘to do’).

The infinitive is used without any preposition after certain verbs expressing causation, mainly caus(e, mak v.1 and gar (as in ModSc ‘gar ye grue’), and also behufe.

Inflected infinitives in -in occur as an anglicism (also spelled -ing).[149]

7.8.10 Present participle

The ending of the present participle in OSc is -and. This is from ON -andi, corresponding to native -inde. In the following example, scraipand is a present participle, sweping is a verbal noun:

Scraipand amang the as be auenture He fand ane iolie iasp, richt precious, Wes castin furth in sweping of the hous (Henryson Fables 68-70).

Adjectives in -ant from French present participles, such as plesand and triumphand, are often written with -and.

By the 16th century, -ing is increasingly replacing -and. The extension of the -ing ending from the verbal noun to the present participle was an innovation in PreStE. The development may, however, be independent in Scots. Where it survives in modern Southern and Northern Scots dialects, -and is pronounced without the /d/. Assuming this to have been general,[150] the ending could easily fall together with the ending of the verbal noun, which, despite its spelling, was commonly pronounced -in (see §6.31.3). But even if we interpret the eclipse of -and as a native development, it is of course reinforced by PreStE usage.

7.8.11 Perfective aspect

Perfective aspect is expressed by the auxiliary verb have + the past participle. After the OE period, have became the dominant auxiliary of the (plu)perfect, perhaps because be was increasingly being used in the passive. A few verbs, mainly of motion, continue to take be, e.g.:

The king ... fyrst into the bate is gane (s.v. ga v. 3 (c)).

The continuing use of be with reflexive verbs has been compared to French usage:

sum ... were Withdrawin thaim in full gret hy (Bruce, quoted by Mustanoja, 1960: 502).

There is no sign in OSc of the Insular Scots generalisation of be as the auxiliary of the perfect throughout.[151]

After modal verbs have is occasionally elided (s.v. have v.1, Additions and Corrections, vol. III). This was probably regular in speech, and is found in the very colloquial Wyf of Auchtirmwchty (Aitken, 1978: 105).

7.8.12 Verb prefixes expressing completion

In OE the idea of completion was expressed by means of verb prefixes, particularly e-, for- and to-, often with the past participle. e survived in sME as i-, and is found as an anglicism in MSc verse (mainly in Douglas, e.g. ibund, yclepit; see y prefix). For-, as in forfochtin ‘exhausted with fighting’, remains an important prefix. To- (q.v.) is sometimes written as a separate word. It is a peculiarity of Douglas to use to- with merely intensive force.

7.8.13 Progressive aspect

Progressive aspect takes the form of the auxiliary verb be + the present participle in -and (or -ing, see §7.8.10).

7.8.14 The passive voice

The passive is constructed with the auxiliary verb be + the past participle. In OE there were two auxiliaries of the passive: wesan/bēon (i.e. be) and weorÞan (which survives into OSc in a range of senses as the main verb worth). The disuse of worth as the auxiliary of the passive after the OE period has been interpreted in terms of Medieval Latin and French reinforcement of be (cf. amatus est, il est aimé, etc.). Passive constructions in general are more characteristic of literary than of colloquial language (see Mustanoja, 1960: 438-40 and 616-19).

Moessner (1997) points also to passives with have and get in the 16th century, e.g.:

Calene, quhilk was Thayn of Anguse and had vij sonnys slayne with King Donald in þis last batall (p.131)
thai gatt ane place gevin to þame be þe King in Pareiß (p.131).

The prepositions introducing the agent in passives are of (q.v. 14), be, and with.

7.8.15 Auxiliary do and gan, can, couth

In EModE, do comes to be used as a tense-carrying auxiliary in interrogative and negative clauses with main verbs (cf. StE ‘do/did they want any?’, ‘no, I don't/didn't think so’). Earlier, do had simply been used in a redundant way in positive indicative clauses, mainly in verse, where it provided an optional extra syllable. This is the way in which it is chiefly used in MSc, mainly at first in Chaucerian-influenced styles of verse (while gan, etc. are similarly used in more native styles):

Pandora sle That with hir slichtis men dois vincous (s.v. do v. B 6 a (2))
I did neuer sleip on Pernaso (ibid. (3))
Whan thay had done full long in armis duell (ibid. b (2)).

Peculiar to Scots is the present participle in this use:

Now mony vsis sosserie, Doand the deuylis of Hell coniure (ibid. (3)).

In Meurman-Solin's corpus of Scots prose, periphrastic do in affirmative declarative sentences first occurs in the 1550s, when it was already at its height in English (prior to a rapid decline in the 17th century), and is most common in the most anglicised texts, suggesting that its use in prose is to some extent referable to intensifying English influence, rather than to its established use in verse (1993: 264). It continues to be common in Scots throughout the period covered by HCOS, i.e. to 1700. See Meurman-Solin (1993a: 267ff.) for a discussion of the factors that favour its use. Do was less frequent as a tense-carrying auxiliary in HCOS, and its textual distribution became increasingly independent of that of periphrastic do over time (1993a: 210, Table XLII).

The redundant auxiliary gan (q.v.) is used only in verse. Originally a reduced form of began (s.v. begin), which was used, though rarely, with the bare infinitive, it became confused with the modal verb can (for discussion, see Markus, 1997).[152] Apart from Barbour and Douglas (who tends towards archaism in other ways), most poets have can rather than gan. The past tense form of can, namely couth, is used in the same way. Both can and couth are treated as past tense forms.

7.8.16 Modal verbs

The modal verbs express hypothetical and conditional states usually expressed in OE simply by the subjunctive.

Modal verbs have only two forms: present tense and preterite. They occur with a main verb (stated or understood):

A bannok is ane good beast, yow may eat the gutis of it on good fryday (s.v. may v.1 5 (a)).

In addition to modals shared with ModStE, such as can/couth, may/micht and will/wald, there are some additional modals:

  • dow/docht ‘be of use, be able to do something’ – this occasionally appears as inflected dowis in the third person singular;
  • man ‘must’, present tense only, from ON;
  • mot, the present tense of must;
  • thar/thur(f)t ‘be obliged’.

Will and sal(l ‘shall’ are still distinct verbs, even in ModSc, and both occur with all persons and numbers. To the extent that will retains some of its original sense of volition, and sal some of its sense of obligation, they are to be regarded as modal rather than merely auxiliary verbs.

There is a putative use of should that is nowadays characteristic of the English of England, but is found with OSc suld(e (s.v. sal(l aux. v. B II 8, 9, 16, 20), whereas ModSc and ScStE reserve should for the sense of moral obligation.[153] Like ModSc and ScStE, OSc also uses wald (s.v. wil(l v.1 34, 36) in such contexts, and also in if clauses (where modern Scottish speech would have no modal, e.g. ‘If you go…’).[154]

There does not appear to be any trace in OSc of double modals (as in ModSc ‘I might could do it’, etc.) (Michael Montgomery, personal communication).

7.8.17 Ellipsis of verbs of motion

The omission of a verb of motion is characteristic of colloquial texts:

Balaam till her again (quoted by Aitken, 1978: 103, from the Pockmanty Preaching)

or with inflection of the prepositional adverb:

he ups with his fist (Aitken, 1978: 103).

7.8.17 Complex verb phrases

During and since the OE period, the verb phrase has become increasingly complex in English and Scots, with various combinations of tense and aspect being expressed periphrastically. This takes place partly under the influence of Latin (and latterly French) prose, and generally as part of the process of elaboration, as the language comes to be used in more formal and scholarly (and therefore more explicit) contexts, replacing Latin. The pluperfect tense, for instance, which has never been much used outside of prose, can be seen as a translation-led development of a native construction.

The perfect and pluperfect (have/had + -en) developed gradually from OE uses of the past participle where the emphasis was on the achieved state, rather than on the action of the verb (cf. ModStE ‘the door is closed’, ‘she has it all arranged’). The simple past tense continued to be used in the OSc period in many contexts where a perfect would now be used.

The be + -and construction developed in OE, apparently under the influence of Medieval Latin (cf. docens erat ‘he was teaching’). Old French also had a periphrastic imperfect (cf. est chantant ‘is singing’). However, its preferred use as the unmarked way of expressing action in the present is a modern development. The simple present was preferred in the OSc period (see further Mustanoja, 1960: 593ff.; Görlach, 1991: 111ff).

The future tense was expressed in OE by the present, with some disambiguating element in the context. The use of wil(l and sal(l ‘shall’ to express futurity was encouraged by the need for explicit translation of Latin future tense forms (see further Mustanoja, 1960: 483, 490). (The modern be going to construction is not found in OSc.)

In the Middle and Modern English periods, more complex forms of the infinitive and passive develop. Again modelled on Latin (and French) is the perfect infinitive, to have + past participle:

Till hawe gottyn ... Of gold and sylvyr gret plente (s.v. have v.1 3 (a)).

The combination of the pluperfect and the progressive is another late development, dating from the mid-16th century in Moessner's data (1997: 114).

(See further Mustanoja, 1960: 516, 518, 525, 527, 551, 553, 554ff., 571 and 574 on Latin influence; 502, 506, 518 and 571 on Old French influence; Görlach, 1991: 110ff.)

7.9 Interrogative pronouns


OE hwelc or hwylc ‘which (of many)’ gives OSc quhilk, with /k/ rather than /ʧ/. The mainly northern /k/ forms may come from the dative form hwelcum, as the change to /ʧ/ did not take place before back vowels, but cf. also the ON cognate hvīlīkr.


OE hwæt, neuter, gives OSc quhat. Quham (from the dative) also survives with non-human reference (s.v. quham pron. 6). In OE, governing the genitive, hwæt could mean ‘how many’, and quhat retains this sense in OSc (s.v. quhat pron. II 10 c).

OE hwa, etc. give quha, quham, quhais. Notice that the dative has again replaced the accusative form.


7.10 Negation

7.10.1 Negative particle

The negative particle negates the verb and therefore the entire clause. In OE it was ne, placed before the verb. This continues in OSc, especially in ESc (see ne adv.1), also taking the nME and Scots form na (see na adv.2), and the anglicised form no (q.v. adv.2). Often another negative, usually nocht (literally ‘nothing’), but sometimes never or some other, follows the verb.

In the course of OSc, nocht (q.v.) becomes the main negative particle. It is normally placed after the verb, though sometimes, mainly in verse, before (see nocht B adv.1 c and 4). (For the do periphrasis, as in ModSc and StE, see §7.8.15.)

Nocht also has reduced forms noth (s.v. nocht), no (q.v. adv.3), and not (q.v.). The last has been treated as an anglicism (Devitt, 1989), but occurs early in Scots, and may be a native development (later reinforced by StE), since reduction is common in unstressed grammatical words.

The enclitic negative particle (= '-n't', following the verb and phonetically run on to it) is na (q.v. adv.4), as in ModSc. It is usually written as a separate word.

The negative adverb ne (q.v. adv.2), which is not recorded as a negative particle in OSc, may or may not be identical with the Modern Northern Scots negative particle nae. SND (s.v. nae adv.2) derives this directly from OE , making it a phonetically conservative form of DOST's na adv.2, although its grammatical behaviour in the modern dialect is like that of no in other Scots dialects, i.e. it comes after a modal or auxiliary verb (including do in support of a main verb). The picture is complicated by the recent reporting of pre-verbal na (of uncertain origin) without do-support in NE Scots (Smith, 2000a, 2000b).

7.10.2 Multiple negation

The small group of words with positive and negative forms are, in OSc: ever/never, or/nor or na, o(w)ther/no(w)thir, ocht/nocht. In the same way, ony and na correspond as positive and negative forms. In OSc the positive forms occur mainly in positive contexts, and it is usual for negatives to occur in negative contexts, giving multiple negation:

Na ky, stirks nor other beists (s.v. na adj. and adv.1 1 b (a))
That we neuir consentit til nane act of parlement ne of counsel (s.v. never adv. 1 b)
Na kyn swylk fundatyown Suld newyre into that plas be set (ibid.)
That na beggar be tholyt to beg nother within burghe nor to lande (s.v. nother B adv. (conj.) 1 c (a)).

However, positive forms do also occur in these contexts:

That na Scottis man suld veir ony clais bot hardyn cotis (s.v. ony 1 adj. b (2)).

7.10.3 Negative adjective

The negative adjective is nan(e (q.v.), or reduced form na (q.v.):

Gret frost that na plwis ȝyd (s.v. na adj. and adv.1 2 (a)).

In ESc, na is only found before consonants, but later also before vowels and /h/.

7.11 Adverbs

The suffix -ly, generally applied in ModStE to adverbs of manner formed from adjectives, is not so regularly applied in Scots (or in dialectal English).

Conversely, OSc sometimes adds the suffix to comparative forms, thus plenarely ‘more plainly’, allanerly ‘uniquely’.

In gretumly ‘greatly’, the adverbial suffix has been added to a dative singular form of the strong adjective.

A characteristic of colloquial texts is the fronting of a prepositional adverb of motion:

incontinent thairefter in comes Thomas Moat (John Campbell's Complaint, quoted by Aitken, 1978: 98).

7.12 Prepositions

Prepositions are used to express a wide range of relationships, many of which were expressed in OE simply by a noun in the appropriate case.

Most prepositions are native, but their range of uses has been often extended in the ME period by the influence of other languages (cf. the periphrastic genitive, §7.4.10) (see further Mustanoja, 1960: 348ff). On the other hand, Görlach (1991: 109) considers that in the EModE period, the influence of Latin has had the effect of restricting the range of uses of the over-worked preposition of, so that it drops out of use, for instance, with the agent in passive sentences (see §7.8.14), while retaining functions where it has the reinforcement of Latin.

New prepositional constructions developed, or were borrowed or calqued, in formal styles, especially:

  • present participles used in absolute constructions (see §7.13.3), e.g. durand (cf. durant, q.v.). Out-takin, out-tane is a literal translation of Latin excepto, Old French excepté. These are not necessarily distinguishable from present participles, thus according (to), for instance, is labelled in DOST as a pres. p.;
  • phrases ending in existing prepositions, e.g. becaus of, instede of, in order to (s.v. order n. 19). These are treated in DOST under the head of the phrase.

(See Mustanoja, 1960: 114-7 and 559-60.)

OE on has become a prefix, reduced to a-, in abone ‘above’, and many similar words (see OED s.v. a-). The reduced form a is also sometimes written for on in OSc, in less polished texts such as local records and letters (see a prep. 1). It was formerly used in English before the verbal noun to express progressive aspect (e.g. a-coming),[155] including a progressive passive, e.g. a-making 'being made'. This is found in OSc only from the late 16th century under English influence (see a prep. 2).

7.13 Subordinate clauses

The elaboration of Scots and English under Latin influence continued a process that had already begun in OE, making it difficult in some cases to distinguish native developments from loans. But, as Sommerfelt remarks:

It seems evident that the elaboration of a literary prose, with a complicated system of subordinate clauses … has taken place in Western and Northern Europe through a general influence from the classical world. We do not know how far the development had gone in the Indo-European Languages of these regions before the classical influence made itself felt. No system of subordinate clauses can be reconstructed for common Indo-European. (1957: 161)

For discussion of the origins of particular constructions, see Blatt (1957), Sørensen (1957).

7.13.1 Relative Pronouns

In OE, Þe was used as an indeclinable relative pronoun. The various parts of se (see §7.5.2) were also used, including Þæt, which gives the indeclinable that of later Scots and English. In ESc, at (q.v.) is usual, from ON at, but is apparently perceived later as a cuttit short form of that and therefore avoided in writing (see §9.3.7). This interpretation is confirmed by the occasional reverse form <that> for at (see that prep.). At survives as the colloquial form of that into ModSc. From the late 16th century on, there are rare citations of quhat as a simple relative pronoun (q.v. V 16).

There is no formal grammatical distinction in ESc between restrictive relative clauses (see quhilk adj. and pron. IV 7) and non-restrictive (descriptive and resumptive, see quhilk adj. and pron. IV 8, 9) relative clauses. Unlike ModStE usage, that occurs freely in non-restrictive use (see that A 1). But by the mid-16th century, the quh- relatives predominate in this function (Romaine, 1982).

A possessive (pro)noun could be relativised, unlike modern usage, where the noun contiguous to the relative pronoun would be understood as its antecedent (see quhilk adj. and pron. IV 9 d).

When a pronominal antecedent is amalgamated with the relative pronoun, quhat can be used as in ModStE, but also that, e.g.:

We sal evin that is od … (s.v. that A 1 c)

also quhilk (q.v. III 5) and rarely quha (q.v. 3). quh- forms

Following the model of Latin and French, the interrogative pronouns are used as relatives in OSc. At first, only quhilk is used, and occurs with human as well as non-human antecedents (see quhilk adj. and pron. IV 7-10). Quha is found as a relative pronoun from the late 16th century, later than quham and quhais (see below). Quham and quhais occur with non-human as well as human antecedents (cf. whose in ModStE) (see quham pron. 6, quhais pron. 6).

At first, the definiteness of the relative as opposed to the interrogative is indicated by adding the, giving the quhilk (cf. Old French liquels) (cf. quhilk adj. and pron. IV 6, 7).

In OSc, the inflected plural form quhilkis is found (see quhilk adj. and pron. IV 7-10), possibly simply replacing the earlier native plural in -e (Mustanoja, 1960: 185), possibly influenced by Latin and French.

The variability of the relative pronouns over time and between genres makes them particularly useful as a rough guide to the character of a text. Case of the relative pronoun

Only the quh- relatives can be marked for case, but the genitive of that could be expressed, as in ModSc, by that + possessive pronoun (see that A 1 d, cf. §7.3.5). The quh- relatives are found earliest in the most complex constructions (Romaine, 1982). From the most to the least likely, these are:

a) genitive (of quhilk, quhais):
For the saulis of al thaim qwhais bodys beis beryst in this kyrk (s.v. quhais pron. 3 a)

b) with other prepositions:
The howse, in the qwilk Ion of Hornedene indwellyt (s.v. quhilk adj. and pron. IV 7 e)

c) direct object (accusative):
Lat thame be victour quham thou lyst avans (s.v. quham pron. 3 c)

d) subject (nominative):
As he the quhilk pretendis to weld the ryng (s.v. quhilk adj. and pron. IV 7 a uninfl. (1)).

In the case of the first two types, the preposition may be postponed, as in ModStE, but in that case, a ‘shadow pronoun’ is used (cf. quhilk adj. and pron. IV 9 l):

Þe scharp croun of thorne, Þe quhilk wes horabill and terrabill to behald the scharp lang pikes of it (Ane Dewoit Exercicioun ll. 130-1, from Bennett ed., 1955). Romance influence

The quh- relatives appeared earliest and were most strongly favoured in prose, rather than verse, under Romance influence. They have never become well established in colloquial Scots (Romaine, 1981).

The quh- pronouns can also be used as adjectives (or ‘determiners’) (see quhilk IV 6). Such dependent relatives have the precedent of Latin grammar. Quhilk as a co-ordinating conjunction

Usage of the quh- relatives was looser than in ModStE. Quhilk (q.v. IV 9 i, j) can relate to the whole of a preceding clause. This construction has the precedent and reinforcement of Latin, but is not especially associated with latinate styles of writing (see Jumpertz-Schwab, 1998: 185ff.). In ModSc, quhilk is obsolete in speech, and this usage is the only one idiomatic for which in colloquial speech. Ø relative

Deletion of the relative pronoun is typical of colloquial ModSc. It appears first in OSc in the nominative, reversing the order of development of the formal quh- relatives (Romaine, 1981):

Here bene the princis, faucht the grete batailis (Kingis Quair §85).

The Ø relative is also possible in the accusative:

the panis ȝe tholit (Ane Dewoit Exercicioun l. 11, from Bennett ed., 1955).

7.13.2 Subordinating conjunctions

The development of a wide range of subordinating conjunctions, expressing various logical relationships, has been part of the elaboration of Scots and English under Latin influence. OE preferred to co-ordinate clauses, a preference which continues in OSc verse and in colloquial language.

Häcker's (1999) study of adverbial clauses in ModSc also touches on the historical development of subordinating conjunctions and their syntax in Scots. With the completion of DOST, it will be possible, together with corpus-based research, to build on her work for a fuller description of OSc, although DOST does not, of course, offer information on constructions with zero conjunction. (When a clause is subordinated to another without any conjunction, the lack of a finite verb signals its subordinate status, as in ModSc and ModStE.)

Häcker, citing Kortmann (1997), outlines the general trend in adverbial subordinators over time: the inventory was continually enlarged between OE and EModE, partly as a result of language contacts, but many of the items added in EModE under Latin influence were short-lived, and there was a general weeding out of diversity from the middle of the 17th century on. Häcker speculates that the EModE phase of expansion would have been less extensive in Scots, with its relative lack of literary activity in the 17th century. Although Häcker has specific examples of conservatism in ModSc, the weeding-out process was extensive here too, and many of the subordinating conjunctions of OSc are unfamiliar to the modern reader, or are used in unfamiliar senses.[156]

That was productive in the ME/OSc period as a general subordination marker, forming the last element of compound conjunctions, e.g. gif that (s.v. gif conj. b), for that (s.v. for prep. 10), quhat that (s.v. quhat pron. III 12 b). This pattern may have been reinforced by French. Häcker suggests on the basis of OED that this use of that "was rarer and was lost earlier in Scots than in Standard English" (1999: 215), but this suggestion remains to be tested. There are numerous other compound conjunctions: to take a few examples almost at random, see for instance insafer, gif na war (s.v. na adv.2 and conj.1 1 b), that na (ibid. 2), nocht-agaynstandand and nocht-obstant, quhat VI 18 conj. b, quhat-tim(e, quhen-swa, and quhidder conj. IV.

On the use of and as a subordinating conjunction with ellipsis of the verb, see §

7.13.3 The absolute participle construction

OSc has an absolute participle construction, consisting of a noun phrase different from the subject of the main clause, plus a participle:

This dovnne the said Gavin sall haf licens (s.v. do v. 3 (e))
Consentand all the barnage hale, Darius tuk the governale (s.v. consent v. 3).

Since the absolute participle construction is already attested in OE, it is not clear whether it was actually borrowed from the Latin ablative absolute, but it is certainly reinforced by Latin. It is particularly favoured by some individuals, notably Bellenden, translating Latin, but also Pitscottie, writing original material in Scots (see Jumpertz-Schwab, 1998: 133ff.).

The identity in form of many preterites and past participles can lead to ambiguity, and it is probably for this reason that be + -and was first introduced into the construction:

and finaly thir two pepill beand distroyitt be this slycht, the Britonis mycht reioiß baith thair realmez in Albioun (quoted by Jumpertz-Schwab, 1998: 140).

Jumpertz-Schwab (1998: 141) points out that the available evidence is insufficient to say whether the construction with beand / being developed first in Scots or in English.

7.13.4 Accusative + infinitive

Another construction encouraged by Latin, though not certainly derived from it, is what Jumpertz-Schwab describes as the infinitive construction after monotransitive verbs, also known as the accusative + infinitive construction:

the pepill … sall never admitt ony persoun regnne abone þame (quoted by Jumpertz-Schwab, 1998: 165)

Tullus … avowit xii preistis … to be perpetualy dedicate to mars (quoted by Jumpertz-Schwab, 1998: 166).

This was found in OE with a narrow range of verbs. It remains very common in OSc, and not stylistically marked, with certain verbs, e.g. gar, but its use was probably expanded under Latin influence.

7.14 Sentence and period

Like modern colloquial speech, OSc can be difficult to parse into sentences, and contemporary punctuation is erratic. The tenses in linked clauses may not be the same, and as Häcker points out (1999: 239), the demarcation between subordination and co-ordination was not so strict in earlier periods, giving amalgamated sentences with characteristics of both. It is sometimes helpful to analyse instead in terms of groups of semantically related clauses, with or without explicit syntactic links.

[114] This section is a revised version of Macafee (1992/93), itself largely based on the standard description of ME grammar, Mustanoja (1960), together with the then available DOST entries on individual words.

Many of the peculiarities of Scots grammar are explained in DOST with reference to their OE sources. Some background information on OE grammar is accordingly given here (see further Campbell, 1959 or any of several textbooks). Some features described below are period features shared with EModE. For a detailed description of the grammar of the latter, see Barber (1997), Görlach (1991).

This brief description concentrates on the characteristics that will be least familiar to a reader used to ModStE, and thus, admittedly, perpetuates the tendency to treat Scots as a series of footnotes to StE. For a treatment of selected aspects of OSc grammar from first principles, see Moessner (1997). Moessner's article, like several others in Jones ed. (1997), would have benefited from reference to DOST.

[115] It has been pointed out that language shift by p-Celtic speakers may already have begun the process of simplification of English morphology (German, 2000: 370).

[116] The isogloss runs through Yorkshire (Kniesza, 1997: 32, citing LALME).

[117] He notes that not all of it is fully reliable, since in the dictionary entries for the earlier part of the alphabet no attention was given to the dates of the actual copies, often much later than those of the (mostly lost) originals. But he accepts the reported forms as evidence in the aggregate.

[118] Apparent early ‘erroneous’ or ‘inorganic’ -e spellings, such as the second -e in <Aldetuneburne> are probably due to later copyists (Aitken, 2002: §13).

[119] Aberd. B. Rec. (SHS).

[120] Facs. Nat. MSS II 14.

[121] The same late ON sound-change gives Shetland from Hjaltland, and Modern SW and Southern Scots choops for (rose)hips, presumably via shoops as in Northern England.

[122] In StE, it develops only at the end of the 16th century.

[123] But Reid in his edition of Hume Douglas notes (p.lix) that it is the normal form in this text.

[124] A sentence quoted by Moessner (1997: 119) does have an example with hir, but this is from a 17th century text, and may reflect English influence.

[125] Apart from Lamb's Resonyng, which is anyway a bilingual English-Scots text.

[126] For a clarification of the pragmatics of these apparent contradictions, see Brown and Levinson (1989).

[127] There is a 1571 example in a sermon quoted by Meurman-Solin (1993: 99).

[128] In the case of my-self and thi-self, the change may have been reinforced by the weakening of the vowel in me and thee, and the change of unstressed e to i, adding to confusion with the genitive forms of the pronouns (OED s.v. myself).

[129] Unlike the nME forms cited by King (1997: 164), the plurals of duke, cace/cas(e, cors n.1 and corps always take the plural inflection in the DOST citations (although corps also gives rise to a new singular corp by misanalysis).

[130] This is the result of a Germanic sound-change known as ‘i-mutation’, from the influence of an i in the following syllable. Changes in the inflections mean that this was not necessarily still present in OE.

[131] With the vowel changed by i-mutation.

[132] Pace King (1997: 165), DOST has no evidence for an uninflected genitive of duke. Uninflected frere occurs in compounds such as freredykis, but DOST regards this as attributive (cf. names such as Gusedubs).

[133] Another instrumental survival is the quantifier quhen(e 'wheen'.

[134] MED also shows some survival of an OE construction with this and that (pronouns) followed by the plural form of the verb be. This would presumably have been over-ridden by the Northern Present Tense Rule (see §7.8.2) had it occurred in OSc.

[135] From Campbell (1959) and OED.

[136] I am grateful to Dr Susanne Kries for providing this list.

[137] On the voicing or not of word-final fricatives, see §6.31.2. Cf. the development of /d/ to /d/ in other environments (see §6.31.4). There may also have been some analogy with secund, thrid and ON-derived forms in -nd.

[138] With a score of 21 for some commodities, the hundred would of course be 126.

[139] These uses are usually entered under the cardinal numbers, but see hundir adj., hundreth(e adj. and (anglicised) eightein(e. There is a coincidence of form between cardinal and ordinal aucht.

[140] Hence the gaps in the record for ordinals too high for measurements of reigns or days of the month and too fine (without being round enough) for fractions. MED also has a few examples of this usage in various contexts, both with figures and words.

[141] WS eall.

[142] She cites a partial precedent in OE.

[143] For an explanation of -is in the third person plural in terms of ON influence (confusion with the ON reflexive ending -sk), see Samuels (1985: 275ff.).

[144] And rarely in the past tense (MacQueen, 1957: 141; Meurman-Solin, 1997b).

[145] If existential sentences, which almost always have is, are excluded (1994: 90).

[146] Jumpertz-Schwab (1998: 200ff.) mistakes present tense bene (an anglicism) in Bellenden for past tense or perfect with elided have (as in §7.8.11).

[147] The percentages reach higher figures than for the present tense (cf. Montgomery's Tables 4 and 5), but the numbers are too low for this to be significant.

[148] Also, probably by confusion with the plural, ar in a 17th century Aberdeen text (s.v. be v. A II 7 (2)); cf. Orkney 1908 (SND s.v. thou).

[149] A reverse spelling from the /ɪn/ pronunciation of -ing (see §6.31.3).

[150] Loss of /d/ after /n/ in stressed syllables is characteristic only of some ModSc dialects. It is general, however, in final unstressed position.

[151] Pavlenko (1997) offers a convincing explanation of this in terms of the decay of Norn, with the falling together of unstressed enclitic forms of the cognates of have and be, reinforced by the residual use of be in Scots.

[152] Pace Moessner (1997: 114) there is no present form gin as a redundant auxiliary.

[153] The terminology is that of Quirk et al. (1972: §§11.22, 32, 72). In other terms, ModSc and ScStE reserve should for deontic modality, whereas in the usage of England it can also be epistemic.

[154] I.e. in the protasis (subordinate clause, typically an if clause) of a conditional sentence. The apodosis is the concluding main clause.

[155] This is a possible source of the English use of –ing for the present participle.

[156] However, some of the conjunctions that are missing from ModSc in Häcker's data, in comparison with ModStE, are found in OSc, e.g. whereas (s.v. quhareas adv. and conj.), and while in NE Scots (s.vv. quhile adv., conj. and prep., quhil(l adv., conj. and prep.).

Macafee, Caroline and †Aitken, A. J. (2002) ‘A history of Scots to 1700’ in A Dictionary of the Older Scottish Tongue vol. XII, xxix-clvii. Online http://dsl.ac.uk/about-scots/history-of-scots/grammar/