The history of the strong verbs traced back through Early OE to PG will reveal the origins of the sound interchanges and of the division into classes; it will also show some features which may help to identify the classes.
The gradation series used in Class 1 through 5 go back to the PIE qualitative ablaut [e–o] and some instances of quantitative ablaut. The grades [e–o] reflected in Germanic as [e/i–a] were used in the first and second stems; they represented the normal grade (a short vowel) and were contrasted to the zero-grade (loss of the gradation vowel) or to the prolonged grade (a long vowel) in the third and fourth stem. The original gradation series split into several series because the gradation vowel was inserted in the root and was combined there with the sounds of the root. Together with them, it was then subjected to regular phonetic changes. Each class of verbs offered a peculiar phonetic environment for the gradation vowels and accordingly transformed the original series into a new gradation series.
In Classes 1 and 2 the root of the verb originally contained [i] and [u] (hence the names i-class and u-class); combination of the gradation vowels with these sounds produced long vowels and diphthongs in the first and second stems. Classes 3, 4 and 5 had no vowels, consequently the first and second forms contain the gradation vowels descending directly from the short [e] and [o]; Class 3 split into subclasses as some of the vowels could be diphthongised under the Early OE breaking. In the third and fourth stems we find the zero-grade or the prolonged grade of ablaut; therefore Class 1 – i-class – has [i]. Class 2— [u] or [o]; in Classes 4 and 5 the Past pl stem has a long vowel [æ]. Class 5 (b) contained [j] following the root in the Inf.; hence the mutated vowel [i] and the lengthening of the consonant: sittan.
In the verbs of Class 6 the original IE gradation was purely quantitative; in PG it was transformed into a quantitative-qualitative series.
Class 7 had acquired its vowel interchange from a different source: originally this was a class of reduplicating verbs, which built their past tense by repeating the root. In OE the roots in the Past tense stems had been contracted and appeared as a single morpheme with a long vowel. The vowels were different with different verbs, as they resulted from the fusion of various root-morphemes, so that Class 7 had no single series of vowel interchanges.
Direct traces of reduplication in OE are rare; they are sometimes found in the Anglian dialects and in poetry as extra consonants appearing in the Past tense forms: Past tense ofhatan — heht alongside het ('call'). Past tense of ondrædan– ondred and ondreord (NE dread).
To account for the interchanges of consonants in the strong verbs one should recall the voicing by Verner's Law and some subsequent changes of voiced and voiceless fricatives. The interchange [s–z] which arose under Verner's Law was transformed into [s–r] due to rhotacism and acquired another interchange [s–z] after the Early OE voicing of fricatives. Consequently, the verbs whose root ended in [s] or [z] could have the following interchange:
ceosan [z] ceos [s] curon[r] coren [r] (NE choose)
Verbs with an interdental fricative have similar variant with voiced and voiceless [0, ð] and the consonant [d], which had developed from [ð] in the process of hardening:
sniþan [ð] snaþ [0] snidonsniden(NE cut)Class1
Verbs with the root ending in [f/v] displayed the usual OE interchange of the voiced and voiceless positional variants of fricatives:
ceorfan [v] cearf [f] curfon [v] corfen [v] (NE carve) Class 3
Verbs with consonant interchanges could belong to any class, provided that they contained a fricative consonant. That does not mean, however, that every verb with a fricative used consonant interchange, for instance risan, a strong verb of Class 1, alternated [s] with [z] but not with [r]: risan – ras – rison – risen (NE rise). Towards the end of the OE period the consonant interchanges disappeared.
Weak Verbs
The number of weak verbs in OE by far exceeded that of strong verbs. In fact, all the verbs, with the exception of the strong verbs and the minor groups (which make a total of about 320 verbs) were weak. Their number was constantly growing since all new verbs derived from other stems were conjugated weak (except derivatives of strong verbs with prefixes). Among the weak verbs there were many derivatives of OE noun and adjective stems and also derivatives of strong verbs built from one of their stems (usually the second stem — Past sg)
talu n – tellan v (NE tale, tell) full adj – fyllan v (NE full, fill)
Weak verbs formed their Past and Participle II by means of the dental suffix -d- or -t- (a specifically Germanic trait). In OE the weak verbs are subdivided into three classes differing in the ending of the Infinitive, the sonority of the suffix, and the sounds preceding the suffix. The main differences between the classes were as follows: in Class I the Infinitive ended in -an, seldom -ian (-ian occurs after [r]); the Past form had -de, -ede or -te; Participle II was marked by –d, -ed or -t. Some verbs of Class I had a double consonant in the Infinitive, others had a vowel interchange in the root, used together with suffixation.
Class II had no subdivisions. In Class II the Infinitive ended in -ian and the Past tense stem and P II had [o] before the dental suffix. This was the most numerous and regular of all the classes.
The verbs of Class III had an Infinitive in -an and no vowel before the dental suffix; it included only four verbs with a full conjugation and a few isolated forms of other verbs. Genetically, the division into classes goes back to the differences between the derivational stem-suffixes used to build the verbs or the nominal stems from which they were derived, and all the persons of the sg Subj. (cf. restan—reste, wendan— wende, (NE rest, wend).
Participle II of most verbs preserved -e- before the dental suffix, though in some groups it was lost.
Several minor groups of verbs can be referred neither to strong nor to weak verbs. The most important group of these verbs were the so-called "preterite-presents" or "past-present" verbs. Originally the Present tense forms of these verbs were Past tense forms (or, more precisely, IE perfect forms, denoting past actions relevant for the "present). Later these forms acquired a present meaning but preserved many formal features of the Past tense. Most of these verbs had new Past Tense forms built with the help of the dental suffix. Some of them also acquired the forms of the verbals: Participles and Infinitives; most verbs did not have a full paradigm and were in this sense "defective".
The verbs were inflected in the Present like the Past tense of strong verbs: the forms of the 1st and 3rd p. sg were identical and had no ending – yet, unlike strong verbs, they had the same root-vowel in all the persons; the pl had a different grade of ablaut similarly with strong verbs (which had two distinct stems for the Past: sg and pl). In the Past the preterite-presents were inflected like weak verbs: the dental suffix plus the endings -e, -est, -e. The new Infinitives sculan, cunnan were derived from the pl form. The interchanges of root-vowels in the sg and pl of the Present tense of preterite-present verbs can be traced to the same gradation series as were used in the strong verbs. Before the shift of meaning and time-reference the would-be preterite-presents were strong verbs. The prototype of can may be referred to Class 3 (with the grades [a–u] in the two Past tense stems); the prototype of sculan — to Class 4, magan — to Class 5, witan, wat 'know' – to Class 1.
In OE there were twelve preterite-present verbs. Six of them have survived in Mod E: OE ag; cunnan, cann; dear(r), sculan, sceal; magan, mæg, mot (NE owe, ought; can; dare; shall; may; must). Most of the preterite-presents did not indicate actions, but expressed a kind of attitude to an action denoted by another verb, an Infinitive, which followed the preterite-present. In other words, they were used like modal verbs, and eventually developed into modem modal verbs. (In OE some of them could also be used as notional verbs:
þe him aht sceoldon 'what they owed him'.)
Among the verbs of the minor groups there were several anomalous verbs with irregular forms. OE willan was an irregular verb with the meaning of volition and desire; it resembled the preterite-presents in meaning and function, as it indicated an attitude to an action and was often followed by an Infinitive.
þa ðe willað mines forsiðes fægnian 'those who wish to rejoice in my death'
hyt moten habban eall 'all could have it'.
Willan had a Past tense form wolde, built like sceolde, the Past tense of the preterite-present sculan, sceal. Eventually willan became a modal verb, like the surviving preterite-presents, and, together with sculan developed into an auxiliary (NE shall, will, should, would).
Some verbs combined the features of weak and strong verbs. OE don formed a weak Past tense with a vowel interchange: and a Participle in -n: don — dyde – gedon (NE do). OE buan 'live' had a weak Past – bude and P II, ending in -n, gebun like a strong verb.
Two OE verbs were suppletive. OE gan, whose Past tense was built from a different root gan – eode – gegan (NE go); and beon (NE be).
Beon is an ancient (IE) suppletive verb. In many languages – Germanic and non-Germanic – its paradigm is made up of several roots. In OE the Present tense forms were different modifications of the roots *wes- and *bhu-, 1st p. sg eom, beo, 2nd p. eart, bist. The Past tense was built from the root *wes-on the pattern of strong verbs of Class 5. Though the Infinitive and Participle II do not occur in the texts, the set of forms can be reconstructed as: *wesan — wæs — wæron — *weren.
OE syntax
The syntactic structure of OE was determined by two major conditions: the nature of OE morphology and the relations between the spoken and the written forms of the language,
OE was largely a synthetic language; it possessed a system of grammatical forms, which could indicate the connection between words; consequently, the functional load of syntactic ways of word connection was relatively small. It was primarily a spoken language, therefore the written forms of the language resembled oral speech – unless the texts were literal translations from Latin or poems with stereotyped constructions. Consequently, the syntax of the sentence was relatively simple; coordination of clauses prevailed over subordination; complicated syntactical constructions were rare.
The syntactic structure of a language can be described at the level of the phrase and at the level of the sentence. In OE texts we find a variety of word phrases (also: word groups or patterns). OE noun patterns, adjective patterns and verb patterns had certain specific features, which are important to note in view of their later changes.
A noun pattern consisted of a noun as the head-word and pronouns, adjectives (including verbal adjectives, or participles), numerals and other nouns as determiners and attributes. Most noun modifiers agreed with the noun in gender, number and case:
on þæm oþrum þrim dagum ... 'in those other three days' – Dat. pl Masc.
Ohthere sæde his hlaforde, Ælfrede cyninge 'Ohthere said to his lord, king Alfred' – the noun in apposition is in the Dat. sg like the head noun.
Nouns, which served as attributes to other nouns, usually had the form of the Gen. case: hwales ban, deora fell 'whale's bone, deer's fell'.
Some numerals governed the nouns they modified so that formally the relations were reversed: tamra deora ... syx hund 'six hundred tame deer'; twentig sceapa 'twenty sheep' (deora, sceapa – Gen. pl).
The following examples show the structure of the simple sentence in OE, its principal and secondary parts:
Soðlice sum mann hæfde twegen suna (mann – subject, hæfde – Simple Predicate) 'truly a certain man had two sons'. Predicates could also be compound: modal, verbal and nominal:
Hwæðre þu meaht singan 'nevertheless you can sing'.
He was swyðe spedig mann 'he was a very rich man'. The secondary parts of the sentence are seen in the same examples: twegen suna 'two sons' – Direct Object with an attribute, spedig 'rich' – attribute. In the examples of verb and noun patterns above we can find other secondary parts of the sentence: indirect and prepositional objects, adverbial modifiers and appositions: hys meder 'to his mother' (Indirect Object), to his suna 'to his son' (Prep. Object), his hlaforde, Ælfrede cyninge 'his lord king Alfred' (apposition). The structure of the OE sentence can be described in terms of Mod E syntactic analysis, for the sentence was made up of the same parts, except that those parts were usually simpler. Attributive groups were short and among the parts of the sentence there were very few-predicative constructions ("syntactical complexes"). Absolute constructions with the noun in the Dat. case were sometimes used in translations from Latin in imitation of the Latin Dativus Absolutus. The objective predicative construction "Accusative with the Infinitive" occurred in original OE texts:
... ða liðende land gesawon, brimclifu blican, beorgas steape (BEOWULF)
'the travellers saw land, the cliffs shine, steep mountains'. Predicative constructions after habban (NE have) contained a Past Participle.
The connection between the parts of the sentence was shown by the form of the words as they had formal markers for gender, case, number and person. As compared with later periods agreement and government played an important role in the word phrase and in the sentence. Accordingly the place of the word in relation to other words was of secondary importance and the order of words was relatively free.
The presence of formal markers made it possible to miss out some parts of the sentence which would be obligatory in an English sentence now. In the following instance the subject is not repeated but the form of the predicate shows that the action is performed by the same person as the preceding action:
þa com he on morgenne to þæm tungerefan se þe his ealdorman wæs; sægde him, hwylce gife he onfeng 'then in the morning he came to the town-sheriff the one that was his alderman; (he) said to him what gift he had received'.
The formal subject was lacking in many impersonal sentences (though it was present in others): Norþan snywde 'it snowed in the North'; him þuhte 'it seemed to him', Hit hagolade stānum 'it hailed with stones'.
One of the conspicuous features of OE syntax was multiple negation within a single sentence or clause. The most common negative particle was he, which was placed before the verb; it was often accompanied by other negative words, mostly naht or noht (which had developed from ne plus awiht 'no thing'). These words reinforced the meaning of negation'.
Ne con ic noht singan... ic noht singan ne cuðe 'I cannot sing' (lit. "cannot sing nothing"), 'I could not sing' (noht was later shortened to not, a new negative particle).
Another peculiarity of OE negation was that the particle ne could be attached to some verbs, pronouns and adverbs to form single words: he ne mihtenan þing geseon 'he could not see anything' (nan from ne an 'not one'), hit na buton gewinne næs 'it was never without war' (næs from ne wæs 'no was'; NE none, never, neither are traces of such forms).
Compound and complex sentences existed in the English language since the earliest times. Even in the oldest texts we find numerous instances of coordination and subordination and a large inventory of subordinate clauses, subject clauses, object clauses, attributive clauses adverbial clauses. And yet many constructions, especially in early original prose, look clumsy, loosely connected, disorderly and wanting precision, which is natural in a language whose written form had only begun to grow.