Trisha Shetty (Editor)

Germanic strong verb

Updated on
Edit
Like
Comment
Share on FacebookTweet on TwitterShare on LinkedInShare on Reddit
Germanic strong verb httpssmediacacheak0pinimgcom736xd5eb37


Similar
  
German pronouns, German conjugation, German adjectives

In the Germanic languages, a strong verb is a verb that marks its past tense by means of changes to the stem vowel (ablaut). The majority of the remaining verbs form the past tense by means of a dental suffix (e.g. -ed in English), and are known as weak verbs.

Contents

In modern English, strong verbs include sing (present I sing, past I sang, past participle I have sung) and drive (present I drive, past I drove, past participle I have driven), as opposed to weak verbs such as open (present I open, past I opened, past participle I have opened). Not all verbs with a change in the stem vowel are strong verbs, however; they may also be irregular weak verbs such as bring, brought, brought or keep, kept, kept. The key distinction is that most strong verbs have their origin in the sound system of Indo-European, whereas weak verbs use a dental ending (in English usually -ed or -t) that was introduced later, during the development of the Germanic languages.

The "strong" vs. "weak" terminology was coined by the German philologist Jacob Grimm, and the terms "strong verb" and "weak verb" are direct translations of the original German terms "starkes Verb" and "schwaches Verb".

Origin and development

Strong verbs have their origin in the ancestral Proto-Indo-European (PIE) language. In PIE, vowel alternations called ablaut were frequent and occurred in many types of word, not only in verbs. The vowel that appeared in any given syllable is called its "grade". In many words, the basic vowel was *e (e-grade), but, depending on what syllable of a word the stress fell on in PIE, this could change to *o (o-grade), or disappear altogether (zero grade). Both e and o could also be lengthened to ē and ō (lengthened grade). Thus ablaut turned short e into the following sounds:

As the Germanic languages developed from PIE, they dramatically altered the Indo-European verbal system. PIE verbs had no tense, but could occur in three distinct "aspects": the present, aorist and perfect aspect. The present implied some attention to such details and was thus used for ongoing actions ("is eating", imperfective aspect). The aorist originally denoted events without any attention to the specifics or ongoing nature of the event ("ate", perfective aspect). The perfect was a stative verb, and referred not to the event itself, but to the state that resulted from the event ("has eaten" or "is/has been eaten"). In Germanic, the aorist eventually disappeared and merged with the present, while the perfect took on a past tense meaning and became a general past tense. The strong Germanic present thus descends from the PIE present, while the past descends from the PIE perfect. The inflexions of PIE verbs also changed considerably.

In the course of these changes, the different root-vowels caused by PIE ablaut became markers of tense. Thus in Germanic, *bʰer- became *beraną in the infinitive (e-grade); *bar in the past singular (o-grade); *bērun in the past plural (ē-grade); and *buranaz in the past participle (zero-grade).

In Proto-Germanic, the system of strong verbs was largely regular. As sound changes took place in the development of Germanic from PIE, the vowels of strong verbs became more varied, but usually in predictable ways, so in most cases all of the principal parts of a strong verb of a given class could be reliably predicted from the infinitive. Thus we can reconstruct Common Germanic as having seven coherent classes of strong verbs. This system continued largely intact in the first attested Germanic languages, notably Gothic, Old English, Old High German and Old Norse.

Gradual disappearance

Germanic strong verbs, mostly deriving directly from PIE, are slowly being supplanted by or transformed into weak verbs.

As well as developing the strong verb system, Germanic also went on to develop two other classes of verbs: the weak verbs and a third, much smaller, class known as the preterite-present verbs, which are continued in the English auxiliary verbs, e.g. can/could, shall/should, may/might, must. Weak verbs originally derived from other types of word in PIE and originally occurred only in the present aspect. They did not have a perfect aspect, meaning that they came to lack a past tense in Germanic once the perfect had become the past. Not having a past tense at all, they obviously also had no vowel alternations between present and past. To compensate for this, a new type of past tense was eventually created for these verbs by adding a -d- or -t- suffix to the stem. This is why only strong verbs have vowel alternations: their past tense forms descend from the original PIE perfect aspect, while the past tense forms of weak verbs were created later.

The development of weak verbs in Germanic meant that the strong verb system ceased to be productive. Practically all new verbs were weak, and few new strong verbs were created. Over time, strong verbs tended to become weak in some languages, so that the total number of strong verbs in the languages was constantly decreasing.

The coherence of the strong verb system is still present in modern German, Dutch, Icelandic and Faroese. For example, in German and Dutch, strong verbs are consistently marked with a past participle in -en, while weak verbs have a past participle in -t in German and -t or -d in Dutch. In English, however, the original regular strong conjugations have largely disintegrated, with the result that in modern English grammar, a distinction between strong and weak verbs is less useful than a distinction between "regular" and "irregular" verbs. Thus the verb to help, which used to be conjugated help-holp-holpen, is now help-helped-helped. The reverse phenomenon, whereby a weak verb becomes strong by analogy, is rare (one example in American English, considered informal by some authorities, is sneak, snuck, snuck. Another is the humorous past tense of "sneeze" which is "snoze").

Some verbs, which might be termed "semi-strong", have formed a weak preterite but retained the strong participle, or rarely vice versa. This type of verb is most common in Dutch:

  • lachen lachte (formerly loech) gelachen ("to laugh")
  • vragen vroeg (formerly vraagde) gevraagd ("to ask")
  • Examples in English are swell, swelled, swollen and (only in American and Canadian dialects) dive, dove, dived.

    Conjugation

    As an example of the conjugation of a strong verb, we may take the Old English class 2 verb bēodan, "to offer" (cf. English "bid").

    This has the following forms:

    While the inflections are more or less regular, the vowel changes in the stem are not predictable without an understanding of the Indo-European ablaut system, and students have to learn four or five "principal parts" by heart – the number depends on whether one considers the third-person singular present tense as a principle part. If we choose to use all five, then for this verb they are bēodan, bīett, bēad, budon, boden. The list of five principal parts includes:

    1. The infinitive: bēodan. The same vowel is used through most of the present tense.
    2. The present tense 3rd singular: bīett. The same vowel is used in the 2nd singular.
    3. The preterite 1st singular (from the PIE perfect): bēad, which is identical to the 3rd singular.
    4. The preterite plural: budon. The same vowel is used in the 2nd singular.
    5. The past participle (from the PIE verbal noun): boden. This vowel is used only in the participle.

    Strictly speaking, in this verb ablaut causes only a threefold distinction: parts 1 and 2 are from the e-grade, part 3 from the o-grade, and parts 4 and 5 from the zero grade. The other two distinctions are caused by different kinds of regressive metaphony: part 2, when it is distinct at all, is always derived from part 1 by Umlaut. In some verbs, part 5 is a discrete ablaut grade, but in this class 2 verb it is derived from part 4 by an a-mutation.

    Strong verb classes

    Germanic strong verbs are commonly divided into 7 classes, based on the type of vowel alternation. This is in turn based mostly on the type of consonants that follow the vowel. The Anglo-Saxon scholar Henry Sweet gave names to the seven classes:

    I. The "drive" conjugation

    II. The "choose" conjugation

    III. The "bind" conjugation

    IV. The "bear" conjugation

    V. The "give" conjugation

    VI. The "shake" conjugation

    VII. The "fall" conjugation

    However, they are normally referred to by numbers alone.

    In Proto-Germanic, the common ancestor of the Germanic languages, the strong verbs were still mostly regular. The classes continued largely intact in Old English and the other older historical Germanic languages: Gothic, Old High German and Old Norse. However, idiosyncrasies of the phonological changes led to a growing number of subgroups. Also, once the ablaut system ceased to be productive, there was a decline in the speakers' awareness of the regularity of the system. That led to anomalous forms and the six big classes lost their cohesion. This process has advanced furthest in English, but in some other modern Germanic languages (such as German), the seven classes are still fairly well preserved and recognisable.

    The reverse process in which anomalies are eliminated and subgroups reunited by the force of analogy is called "levelling", and it can be seen at various points in the history of the verb classes.

    In the later Middle Ages, German, Dutch and English eliminated a great part of the old distinction between the vowels of the singular and plural preterite forms. The new uniform preterite could be based on the vowel of the old preterite singular, on the old plural, or sometimes on the participle. In English, the distinction remains in the verb "to be": I was, we were. In Dutch, it remains in the verbs of classes 4 & 5 but only in vowel length: ik brak (I broke - short a), wij braken (we broke - long ā). In German and Dutch it also remains in the present tense of the preterite presents. In Limburgish there is a little more left. E.g. the preterite of to help is (weer) hólpe for the plural but either (ich) halp or (ich) hólp for the singular.

    In the process of development of English, numerous sound changes and analogical developments have fragmented the classes to the extent that most of them no longer have any coherence: only classes 1, 3 and 4 still have significant subclasses that follow uniform patterns.

    For a treatment of the classes one at a time showing how the forms evolved in the various Germanic languages, one can consult an older version of this Wikipedia article.

    Before looking at the seven classes individually, the general developments that affected all of them will be noted. The following phonological changes that occurred between Proto-Indo-European and Proto-Germanic are relevant for the discussion of the ablaut system.

  • The development of grammatischer Wechsel as a result of Verner's law (the voicing of fricatives after an unstressed vowel). This created variations in the consonant following the ablaut vowel.
  • When the zero grade appears before l, r, m or n, the vowel u was inserted, effectively creating a new "u-grade".
  • oa (also oyai, owau).
  • ei when i, ī or j followed in the next syllable. This change is known as umlaut, and was extended to affect other vowels in most later languages.
  • eyī as a result of the above.
  • ei before m or n followed by another consonant. This had the effect of splitting class 3 into 3a and 3b.
  • For the purpose of explanation, the different verb forms can be grouped by the vowel they receive, and given a "principal part" number:

    1. All forms of the present tense, including the indicative mood, subjunctive mood, imperative mood, the infinitive and present participle.
    2. The 1st person and 3rd person singular forms of the past tense in the indicative mood.
    3. All other past tense forms, which includes the past dual and plural in the indicative mood, and all forms of the past subjunctive mood.
    4. The past participle, alone.

    Classes 1 to 6

    The first 5 classes appear to continue the following PIE ablaut grades:

    Except for the apparent ē-grade in part 3 of classes 4 and 5, these are in fact straightforward survivals of the PIE situation.

    The standard pattern of PIE is represented in Germanic by classes 1, 2 and 3, with the present (part 1) in the e-grade, past indicative singular (part 2) in the o-grade, and remaining past (part 3) and past participle (part 4) in the zero grade. The differences between classes 1, 2, and 3 arise from semivowels coming after the root vowel, as shown in the table below.

    As can be seen, the e-grade in part 1 and o-grade in part 2 are shared by all of these five classes. The difference between them is in parts 3 and 4:

  • In classes 1 and 2, the semivowel following the vowel was converted in the zero grade into a full vowel.
  • In class 3 and the past participle of class 4, there was no semivowel but there were PIE syllabic resonants which developed in Germanic to u plus resonant; thus u became the Germanic sign of these parts. There is some evidence that this may have been the original behaviour of the past nonsingular / nonindicative of class 4 as well: to wit, preterite-presents whose roots have the class 4 shape show u outside the present indicative singular, such as *man- ~ *mun- "to remember", *skal- ~ *skul- "to owe".
  • In class 5, the zero grade of the past participle had probably been changed to e-grade already in PIE, because these verbs had combinations of consonants that were phonotactically illicit as a word-initial cluster, as they would be in the zero grade.
  • The *ē in part 3 of classes 4 and 5 is not in fact a PIE lengthened grade but arose in Germanic. Ringe suggests that it was analogically generalised from the inherited part 3 of the verb *etaną "to eat" before it had lost its reduplicant syllable, PIE *h1eh1d- regularly becoming Germanic *ēt-.
  • Class 6 appears in Germanic with the vowels a and ō. PIE sources of the a vowel included *h2e, *o, and a laryngeal between consonants; possibly in some cases the a may be an example of the a-grade of ablaut, though the existence of such a grade is controversial. It is not clear exactly how the ō is to be derived from an earlier ablaut alternant in PIE, but believable sources include contraction of the reduplicant syllable in PIE *h2-initial verbs, or o-grades of verbs with interconsonantal laryngeal. In any event, within Germanic the resulting a ~ ō behaved as just another type of vowel alternation.

    In Proto-Germanic, this resulted in the following vowel patterns:

  • Class 2b is of unknown origin, and does not seem to reflect any PIE ablaut pattern.
  • In class 3, there are also a few cases where the vowel is followed, at least in Proto-Germanic, by two consonants, neither of which is a nasal or a liquid. Examples: *brestaną "to burst", *þreskaną "to thresh" *fehtaną "to fight". All but one have a nasal or a liquid in front of the vowel. This will have become syllabic and resulted regularly in u before the nasal or liquid, which was then metathesised on the analogy of the remaining principle parts. E.g. part 3 of *brestaną will have been *bʰr̥st- > *burst-, reformed to *brust-.
  • Similarly, class 6 includes some cases where the vowel is followed by two obstruents, like *wahsijaną "to grow".
  • In classes 5, 6 and 7, there is also a small subgroup called "j-presents". These form their present tense with an extra -j-, which causes umlaut in the present where possible. In West Germanic, it also causes the West Germanic gemination.
  • Class 7

    The forms of class 7 were very different and did not neatly reflect the standard ablaut grades found in the first 5 classes. Instead of (or in addition to) vowel alternations, this class displayed reduplication of the first consonants of the stem in the past tense.

    It is generally believed that reduplication was once a feature of all Proto-Indo-European perfect-aspect forms. It was then lost in most verbs by Proto-Germanic times due to haplology. However, verbs with vowels that did not fit in the existing pattern of alternation retained their reduplication. Class 7 is thus not really one class, but can be split into several subclasses based on the original structure of the root, much like the first 5 classes. The first three subclasses are parallel with classes 1 to 3 but with e replaced with a: 7a is parallel to class 1, class 7b to class 2, and class 7c to class 3.

    The following is a general picture of the Proto-Germanic situation as reconstructed by Jay Jasanoff. Earlier reconstructions of the 7th class were generally based mostly on Gothic evidence.

    The situation sketched above did not survive intact into any of the Germanic languages. It was changed significantly, but rather differently in Gothic on one hand, and in the Northwest Germanic languages on the other.

    Gothic

    Reduplication was retained in Gothic, with the vowel ai inserted. However, as in all other strong verbs, consonant alternations were almost entirely eliminated in favour of the voiceless alternants. The present and past singular stem was extended to the plural, leaving the reduplication as the only change in the stem between the two tenses. The vowel alternation was retained in a few class 7d verbs, but eliminated otherwise by generalising the present tense stem throughout the paradigm. The verb lētan "to allow" retained the past form lailōt with ablaut, while slēpan "to sleep" had the past tense form saislēp without it. The form saizlēp, with Verner-law alternation, is occasionally found as well, but it was apparently a relic formation with no other examples of alternation elsewhere.

    Northwest Germanic

    In the Northwest Germanic languages, which include all modern surviving Germanic languages, class 7 was drastically remodelled. Reduplication was almost eliminated, except for a few relics, and new ablaut patterns were introduced. Many attempts were made to explain this development. Jasanoff posits the following series of events within the history of Northwest Germanic:

    1. Root-initial consonant clusters were transferred to the beginning of the reduplicating syllable, to preserve the same word onset across the paradigm. The clusters were simplified and reduced medially. (Compare Latin scindō ~ scicidī and spondeō ~ spopondī, which show the same development)*hlaupaną: *hehlaup, *hehlupun > *hlelaup, *hlelupun*stautaną: *stestaut, *stestutun > *stezaut, *stezutun*blōtaną: *beblōt, *beblutun > *blelōt, *blelutun*grōaną: *gegrō, *gegrōun > *grerō, *grerōun*swōganą: *sezwōg, *sezwōgun > *swewōg, *sweugun (English sough)
    2. Root compression:
      1. Based on the pattern of verbs such as singular *lelōt, *rerōd ~ plural *leltun, *rerdun, as well as verbs like singular *swewōg ~ plural *sweugun, the root vowel or diphthong was deleted in the past plural stem. The Germanic spirant law caused devoicing in certain consonants where applicable.*haitaną: *hegait, *hegitun > *hegait, *hehtun*bautaną: *bebaut, *bebutun > *bebaut, *beftun ("to beat")*hlaupaną: *hlelaup, *hlelupun > *hlelaup, *hlelpun*stautaną: *stezaut, *stezutun > *stezaut, *stestun*blōtaną: *blelōt, *blelutun > *blelōt, *bleltun
      2. In class 7c verbs, this resulted in consonant clusters that were not permissible (e.g. **hegldun); these clusters were simplified by dropping the root-initial consonant(s).*haldaną: *hegald, *heguldun > *hegald, *heldun*fanhaną: *febanh, *febungun > *febanh, *fengun
    3. The present plural stem of class 7c verbs no longer appeared to be reduplicated because of the above change, and was extended to the singular. This created what appeared to be a new form of ablaut, with a in the present and e in the past plural.*haldaną: *hegald, *heldun > *held, *heldun*fanhaną: *febanh, *fengun > *feng, *fengun
    4. This new form of ablaut was then extended to other classes, by alternating *a with *e in classes 7a and 7b, and *ā with *ē in class 7d (after Proto-Germanic *ē had become *ā in Northwest Germanic). In class 7a, this resulted in the vowel *ei, which soon merged with *ē (from Germanic *ē2).*haitaną: *hegait, *hehtun > *heit, *heitun > *hēt, *hētun*hlaupaną: *hlelaup, *hlelpun > *hleup, *hleupun*lātaną: *lelōt, *leltun > *lēt, *lētun
    5. It is at this point that North and West Germanic begin to diverge.
    6. In West Germanic, class 7e took *eu as the past stem vowel, by analogy with existing verbs with initial *(s)w- such as *wōpijaną, *weup(un) and *swōganą, *swewg(un).
    7. In North Germanic, class 7e instead took *ē as the past stem vowel, probably by analogy with class 7c which also had a long stem vowel.

    Stages 4 and 5 were not quite complete by the time of the earliest written records. While most class 7 verbs had replaced reduplication with ablaut entirely, several vestigial remains of reduplication are found throughout the North and West Germanic languages. Various other changes occurred later in the individual languages. *e in class 7c was replaced by *ē (> ia) in Old High German and Old Dutch, but by *eu (> ēo) in Old English.

    The following "Late Proto-Northwest-Germanic" can be reconstructed as descendants of the earlier Proto-Germanic forms given above. Note that ē became ā in Northwest Germanic.

    Proto-Germanic

    The Proto-Germanic language had about 500 strong roots. Although some roots are speculative, the language can be reconstructed with the following strong roots. Proto-Germanic had aorist-present roots, a remnant of the aorist aspect found in Proto-Indo-European. These verbs used the former aorist as a present tense form. The aorist had a reduced vowel, like parts 3 and 4 of the perfect. So these verbs have an anomalous vowel in the present tense, they decline regularly otherwise.

  • Class 1 with about 102 roots: *bīdaną, *bītaną, *blīkaną, *dīganą, *dīkaną, *drībaną, *drītaną, *dwīnaną, *flītaną, *fīganą, *fīsaną, *glīaną, *glīdaną, *glīmaną, *glīnaną, *glītaną, *gnīdaną, *grīnaną, *grīpaną, *hnīganą, *hnīgwaną, *hnītaną, *hnīwaną, *hrībaną, *hrīnaną, *hrītaną, *hwīnaną, *kīkaną, *kīnaną, *klībaną, *knīpaną, *krīganą, *kwīnaną, *krīaną, *krītaną, *kwīþaną, *lībaną, *līhwaną, *līþaną, *mīganą, *mīþaną, *nīwaną, *pīpaną, *rīdaną, *rīfaną, *rīhaną, *rīkaną, *rīmaną, *rīpaną, *rīsaną, *rīstaną *sīganą, *sīhwaną, *sīpaną, *sīþaną, *skīdaną, *skīnaną, *skītaną, *skrīaną, *skrībaną, *skrīhaną, *skrītaną, *skrīþaną, *slībaną, *slīkaną, *slīpaną, *slītaną, *smītaną, *snīkaną, *snīþaną, *snīwaną, *spīwaną, *splītaną, *sprīþaną, *stīganą, *strībaną, *strīdaną, *strīkaną, *strīmaną, *swībaną, *swīganą, *swīkaną, *swīkwaną, *swīnaną, *swīþaną, *tīhaną, *þīhaną, *þīnaną, *þrībaną, *þrīnaną, *wīhaną, *wīkwaną, *wīpaną, *wītaną, *wlītaną, *wrībaną, *wrīhaną, *wrītaną, *wrīþaną;
  • Aorist-present roots: *diganą, *stikaną, *wiganą;
  • Class 2a with 53 roots: *beudaną, *beuganą, *bewwaną, *blewwaną, *breutaną, *breuþaną, *brewwaną, *dreuganą, *dreusaną, *feukaną, *fleuganą, *fleuhaną, *fleutaną, *fneusaną, *freusaną, *geusaną, *geutaną, *greutaną, *hleutaną, *hneupaną, *hneusaną, *hnewwaną, *hrewwaną, *keusaną, *kewwaną, *kleubaną, *kreukaną, *kreupaną, *leudaną, *leuganą, *leukaną, *leusaną, *leustaną, *neutaną, *reudaną, *reufaną, *reukaną, *reutaną, *seukaną, *seuþaną, *skeubaną, *skeutaną, *sleutaną, *smeuganą, *smeukaną, *snewwaną, *spreutaną, *steubaną, *streukaną, *streupaną, *teuhaną, *þleuhaną, *þreutaną;
  • Class 2b with 10 roots: *dūkaną, *lūkaną, *lūtaną, *slūkaną, *skūlaną, *smūganą, *sprūtaną, *sūganą, *sūkaną, *sūpaną;
  • Class 3a with 65 roots: *biginnaną, *bindaną, *brinnaną, *dimbaną, *dingwaną, *dintaną, *drinkaną, *finþaną, *grimmaną, *grindaną, *hinkaną, *hinþaną, *hlimmaną, *hrindaną, *kimbaną, *klimbaną, *klimpaną, *klinganą, *klinkwaną, *krimpaną, *kringaną, *krintaną, *kwinkaną, *limpaną, *lingwaną, *linnaną, *ninþaną, *rinnaną, *singwaną, *sinkwaną, *sinnaną, *sinþaną, *skrimpaną, *skrindaną, *skrinkwaną, *slimpaną, *slindaną, *slingwaną, *slinkaną, *slintaną, *spinnaną, *springaną, *sprintaną, *stinkwaną, *swimmaną, *swindaną, *swinganą, *swinkaną, *tinganą, *tinhaną, *tinnaną, *tinþaną, *trimpaną, *trinnaną, *þindaną, *þinhaną, *þinsaną, *þrimmaną, *þrinhaną, *þrinhwaną, *þrintaną, *þwinganą, *windaną, *winnaną, *wringaną,;
  • Class 3b with 63 roots: *belganą, *berganą, *berkaną, *bregdaną, *brestaną, *delbaną, *derbaną, *fehtaną, *felhaną, *fertaną, *flehtaną, *fregnaną, *geldaną, *gellaną, *gelpaną, *gerdaną, *gerraną, *gnellaną, *gnestaną, *hellaną, *helpaną, *hneskwaną, *hrespaną, *hwelbaną, *hwerbaną, *kerbaną, *kwerraną, *melkaną, *meltaną, *rewtaną, *serþaną, *skelbaną, *skellaną, *skerfaną, *skerpaną, *skertaną, *smellaną, *smeltaną, *smertaną, *snerhaną, *snerkaną, *snerpaną, *snertaną, *spurnaną (aorist), *sterbaną, *swelganą, *swellaną, *sweltaną, *swerbaną, *swerkaną, *teldaną, *þehsaną, *þersaną, *þreskaną, *þwersaną, *weltaną, *welwaną, *werpaną, *werraną, *wersaną, *werzaną, *werþaną, *wreskwaną;
  • Class 4 with 24 roots: *beraną, *brekaną, *bremaną, *dwelaną, *felaną, *helaną, *hlemaną, *klenaną, *knudaną (aorist), *kwelaną, *kwemaną/ *kumaną (aorist), *kweraną, *nemaną, *skeraną, *stelaną, *stenaną, *swemaną, *sweraną, *temaną, *teraną, *tremaną, *trudaną (aorist), *þweraną, *welaną/ *wulaną (aorist);
  • Class 5 with 40 roots: dewaną, drepaną, esaną, etaną, fetaną, fregnaną/ frehnaną, gawedaną, gebaną, getaną, hlefaną, hwekaną, jehaną, jesaną, knedaną, kweþaną, lesaną, metaną, nesaną, plehaną, rekaną, rekwaną, sehwaną, skehaną, skrekkaną, sprekaną, swefaną/ sufaną*, trekaną, tredaną, treganą, þrekaną, webaną, wedaną, weganą, wesaną, wrekaną;
  • Class 6 with 42 roots: *akaną, *alaną, *ananą, *bakaną, *dabaną, *daujaną, *draganą, *faraną, *flahaną, *galaną, *gaujaną, *gnaganą, *grabaną, *hlaþaną, *hnabaną, *kalaną, *klaujaną, *lahaną, *lapaną, *malaną, *sakaną, *skabaną, *skakaną, *slahaną, *snakaną, *spananą, *standaną, *takaną, *wadaną, *þwahaną, *wakaną, *wahwnaną;
  • Class 7
  • 7a with 10 roots: *aihaną, *aikaną, *fraisaną, *haitaną, *laikaną, *maitaną, *skaidaną/ *skaiþaną, *swaipaną, *taisaną, *þlaihaną;7b with 11 roots: *audaną, *aukaną, *ausaną, *bautaną, *brautaną, *dauganą, *haufaną, *hlaupaną, *naupaną, *skraudaną, *stautaną;7c with 24 roots: *arjaną, *bannaną, *blandaną, *dawjaną, *faldaną/ *falþaną, *falganą, *fallaną, *faltaną, *fanhaną, *haldaną, *ganganą, *hanhaną, *hawwaną, *saltaną, *skaldaną, *spaldaną, *spannaną, *staldaną, *stanganą, *waldaną, *walkaną, *wallaną, *waltaną, *waskaną;7d with 23 roots: *bēganą, *blēaną, *blēsaną, *brēaną, *brēdaną, *dēaną, *drēdaną, *gēaną, *grētaną, *hwētaną, *knēaną, *krēaną, *lējaną, *lētaną, *mēaną, *nēaną, *rēdaną, *sēaną, *slēpaną, *stēaną, *tēkaną, *þrēaną, *wēaną;7e with 22 roots: *blōaną, *blōtaną, *bnōwwaną, *bōwwaną, *brōaną, *brōkaną, *dōnaną, *flōaną, *flōkaną, *glōaną, *grōaną, *hlōaną, *hrōpaną, *hwōpaną, *hwōsaną, *knōdaną, *rōaną, *snōwaną, *spōaną, *swōganą, *þrōwaną, *wōpijaną;

    Gothic

    Being the oldest Germanic language with any significant literature, it is not surprising that Gothic preserves the strong verbs best. However, some changes still occurred:

  • e > i, eliminating the distinction between the two vowels.
  • i > e (spelled <ai>) and u > o (spelled <au>) when followed by r, h or ƕ.
  • Consonant alternations are almost entirely eliminated by generalising the voiceless alternant across all forms.
  • Also, long ī was spelled <ei> in Gothic.

    Note: The letter transliterated as q was probably a voiceless labiovelar stop, /kʷ/ ([kʷʰ]), comparable to the Latin qu.

    West Germanic

    Changes that occurred in the West Germanic languages:

  • ē > ā
  • a-mutation: u > o when a follows in the next syllable. This affected the past participles of classes 2-4. However, an intervening m or n + consonant blocked this, so the past participle of class 3a kept u.
  • Extension of umlaut to back vowels, causing it to apply also to verbs of class 6.
  • The perfective prefix ga- came to be used (but neither exclusively nor invariably) as a marker of the participle. In English this prefix disappeared again in the Middle Ages.
  • English

    The following changes occurred from West Germanic to Old English:

  • ai > ā
  • eu > ēo
  • au > ēa
  • a > æ except when a back vowel followed in the next syllable
  • ā > ǣ
  • Breaking before certain consonants: æ > ea and e > eo
  • West Saxon Palatalisation: i > ie after g
  • The following are the paradigms for Old English:

    With j-presents (and other anomalies):

  • hebban hōf hōfon hafen ("to raise, heave")
  • scieppan scōp scōpon scapen ("to create, shape")
  • swerian swōr swōron sworen ("to swear")
  • The verb "to stand" follows class 6, but has an anomalous loss of its -n- in the past:

  • standan stōd stōdon standen
  • Some relics of class 7 reduplication remain in Old English, mostly in texts from Anglia (infinitive and past singular shown):

  • bēatan beoft ("to beat")
  • hātan hēht ("to call")
  • lācan leolc ("to move about, leap")
  • lǣtan leort ("to let")
  • on-drǣdan on-dreord ("to dread")
  • rēdan reord ("to advise")
  • spātan speoft ("to spit")
  • Changes that occurred from Old English to Modern English:

  • ā > ō
  • Great Vowel Shift
  • The old second-person singular ("thou") form acquires the ending "-st" in the past, but the second-person singular falls out of common use and is replaced with the second-person plural.
  • Elimination of almost all verb inflection in strong verbs, except for the third-person singular present ending -s (and the second-person ending "-(e)st", when used).
  • Either the past singular form or the past plural form is generalised to the other number. As a result, only one form exists for all past tense forms and parts 2 and 3 are no longer distinguished.
  • Combined with the above, all consonant alternations are eliminated by generalising the consonant of the present. Only be preserves the alternation: was versus were.
  • In Modern English, generally speaking, the verb classes have disintegrated and are not easily recognisable.
    For the principal parts of all English strong verbs see: Wiktionary appendix: Irregular English verbs.

    The following modern English verbs resemble the original paradigm:

    Class 1

    Class 1 is still recognisable, as in most other Germanic languages. The modern past is taken from either the old past singular (ride rode ridden) or the old past plural (bite bit bitten). In the case of shine shone shone, the past participle has also assimilated to the past singular.

    Class 1 verbs in modern English (excluding derived verbs such as abide and override) are bide, bite, chide, drive, hide, ride, rise, rive, shine, shrive, slide, smite, stride, strike, strive, thrive, write.

    However, although these verbs have uniformity in their infinitive vowel, they no longer form a coherent class in further inflected forms – for example, bite (bit, bitten), ride (rode, ridden), shine (shone, shone), and strike (struck, struck/stricken, with struck and stricken used in different meanings) all show different patterns from one another – but bide, drive, ride, rise, smite, stride, strive, write do form a (more or less) coherent subclass. Most of these verbs are descended from Old English class 1 verbs. However:

  • strive is a French loan-word which is class 1 by analogy to drive. (By coincidence it is ultimately descended from an Old Frankish class 1 verb.)
  • thrive is a class 1 verb formed by analogy to drive, its Old English ancestor being weak and descended from Old Norse þrífa (itself a class 1 strong verb, meaning "to grasp").
  • hide is a class 1 verb whose Old English ancestor, hȳdan, was weak.
  • In American English, the past tense of the verb dive is usually dove, as though it is in Class 1, but the past participle is still dived.

    Class 2

    Class 2 has become a small group and has become rather irregular. It includes choose, cleave, fly, freeze and shoot (whose usual passive participle is shot rather than shotten). The verb bid (in the sense of "to offer") was in Class 2, but now the past and ast participle are bid. The obsolete verb forlese is now used only as the passive participle forlorn. This group does not form a coherent class, as each verb has different irregularities from each other verb.

    Class 3

    Class 3 in English is still fairly large and regular. The past is formed either from the old past singular or from the past plural. Many of the verbs have two past forms, one of which may be dialectal or archaic (begin, drink, ring, shrink, sing, slink, spring, stink, swing, swim and wring). However, there are some anomalies. The class 3 verbs in modern English are:

  • With nasal, normal past in a: begin, drink, ring, run, shrink, sing, sink, spring, stink, swim, wring (wring may have either past form)
  • With nasal, normal past in u or ou: bind, cling, find, fling, grind, sling, slink, spin, sting, string, swing, win, wind, wring
  • With ll: swell (but past is now swelled)
  • With original "Germanic h": fight
  • English fling does not go back to Old English, and may be a loan-word from Norse. It seems to have adopted class 3 forms by analogy with cling etc. Similarly ring, string.

    Class 4

    The verb come is anomalous in all the West Germanic languages because it originally began with qu-, and the subsequent loss of the w sound coloured the vowel of the present stem.

  • cuman cymþ cōm cōmon cumen ("to come")
  • Also anomalous:

  • niman nimþ nōm nōmon numen ("to take")
  • In Modern English, regular class 4 verbs have all kept the –n in the participle, though eliminating the medial e after r, this class exhibits near homogeneity of vowel pattern:

  • break broke broken
  • but several verbs have archaic preterites that preserve the "a" of Middle English (bare, brake, gat, sware, tare, and spake or Scots spak). The preterite of Middle English comen was either cam or com (or with -en in the plural).

    Class 4 verbs in English (not including derivatives such as beget) are bear, break, get, shear, speak, steal, swear, tear, tread, wake, weave; and without the -n and of irregular vowel progression: come. Get, speak, tread and weave were originally of class 5, whereas swear was originally class 6. Wake was also originally class 6, and in fact retains the "a" of the present tense – the preterite woke (Middle English wook) only conforms to the modern class 4 preterite, not to the historic class 4 preterite in "a".

    Class 5

    In Modern English this group has lost all group cohesion.

  • eat ate eaten
  • give gave given
  • lie lay lain
  • see saw seen
  • sit sat sat
  • Class 5 verbs in Modern English: bid (in the sense of "to command" or "to invite"), eat, give, lie (= lie down), see, sit. The verb quethe is only used poetically now. Get, speak, tread, weave are now class 4.

    The preterite of the verb forbid can be forbad or forbade, and the preterite ate is pronounced "et" in some British dialects.

    Although the verb to be is suppletive and highly irregular, its past follows the pattern of a class 5 strong verb, with grammatischer Wechsel (the alternation of "s" and "r" in "was" versus "were"), and has uniquely retained the singular/plural distinction of both ablaut grade and consonant in the modern languages. Old English: wæs/wǣron, English: was/were. For full paradigms and historical explanations see Indo-European copula.

    Class 6

    Class 6 has disintegrated as well. The verbs shake, take and forsake come closest to the original vowel sequence. The consonant anomaly in stand is still visible, and is extended to the participle.

  • shake shook shaken
  • stand stood stood
  • Class 6 verbs in modern English: draw, forsake, lade, shake, shape, shave, slay, stand, take. The verb heave is in this class when used in a nautical context. Like most other classes in Modern English, this class has lost cohesion and now forms principal parts according to many different patterns. Two preterites (drew and slew) are now spelled with "ew", which is similar in sound to the "oo" of the others that still use a strong form. Swear is now class 4. The adjective graven was originally a past participle of the now obsolete verb grave. Note that lade, shape, shave, wax are now weak outside of their optionally strong past participle forms (laden, shapen, shaven, and waxen respectively). Fare has archaic past tense fore and rare past participle faren, but is normally weak now.

    Class 7

    In Modern English this class has lost its homogeneity:

  • fall fell fallen
  • hang hung hung (Note that, in the transitive sense of hanging someone by the neck, hang usually has regular weak conjugation hanged)
  • hold held held (the original past participle is preserved in the adjective beholden)
  • throw threw thrown
  • The following modern English verbs descend from class 7 verbs, and still retain strong-verb endings: beat, blow, fall, hew, grow, hang, hold, know, throw. (Hew can be a preterite or present, although the usual preterite is hewed.) The verb let can be considered Class 7, though the past participle now lacks the ending -en. The verbs mow and sow retain the strong-verb participles mown and sown but the preterites are now mowed and sowed. (The verb sew was always weak, even though one can say sewn for the past participle.) Archaic English still retains the reduplicated form hight ("called", originally a past tense, usually with a passive meaning, but later also used as a passive participle). The verb crow was also in class 7, as in the King James Version "while he yet spake, the cock crew".

    Other

    A few English verbs have become "strong" but do not belong to any class:

  • dig, dug, dug
  • dive, dove, dived (Only in American English. Like Class 1 in first two forms.)
  • sneak, snuck, snuck (Non-standard American English. Sneaked is also still used.)
  • spit, spat, spat (Spit, spit, spit is also used.)
  • stick, stuck, stuck
  • strike, struck, struck (Strike is also Class 1 but with an unusual past: strike, struck, stricken.)
  • Dutch

    Old Dutch is attested only fragmentarily, so it is not easy to give forms for all classes. Hence, Middle Dutch is shown here in that role instead. The situation of Old Dutch generally resembled that of Old Saxon and Old High German in any case.

    Changes from West Germanic to Old Dutch:

  • ai > ē (but sometimes ei is preserved)
  • au > ō
  • eu > iu
  • ē > ie
  • ō > uo (later becomes /uə/, spelled <oe> in Middle Dutch)
  • From Old Dutch to Middle Dutch:

  • u > o
  • ū > ȳ (spelled <uu>)
  • iu > ȳ (southern dialects)
  • iu > io > ie (northern dialects)
  • Lengthening of vowels in open syllables: e > ē, o > ō, a > ā, although it continues to be written with a single vowel. i is lengthened to ē, and short y (from umlaut of u) to ue /øː/.
  • Unlike most other languages, umlaut does not affect long vowels or diphthongs except in the eastern dialects.
  • Because of the combined effect of the two above points, umlaut is eliminated as a factor in verb conjugation.
  • From Middle Dutch to Modern Dutch:

  • Diphthongisation of long high vowels: /iː/ > /ɛi/, /yː/ > /œy/ (spelled <ij> and <ui>)
  • Monophthongisation of opening diphthongs: /iə/ > /i/, /uə/ > /u/ (still spelled <ie> and <oe>)
  • Class 1

    This class is well preserved and has the most strong verbs.

  • Historical class 1 strong roots: bezwijken, bijten, blijken, blijven, drijten (dialect), drijven, glijden, grijpen, kijken, knijpen, krijgen, krijten, lijden, lijken, mijden, mijgen/ miegen (dialect), nijgen, nijpen, pijpen (in the sense playing a flute), rijden, rijgen, rijten, rijven, rijzen, schijnen, schijten, schrijden, schrijven, slijpen, slijten, smijten, snijden, splijten, stijgen, strijden, strijken, (aan)tijgen, verdwijnen, wijken, wijten, wrijten, wrijven, zijgen, zwijgen.
  • Historically weak Class 1 roots: hijsen, kijven, krijsen, kwijten, prijzen, spijten, stijven, vijzen, vrijen, wijzen
  • Three verbs with 'ei' have also joined this class by analogy, as 'ij' and 'ei' are pronounced the same: breien, uitscheiden, zeiken.
  • Class 2

    Class 2b has grown greatly by moving older class 2a verbs into it. In most dialects the historical "eu" merged with "ū", as both developed into "ȳ" and later into "ui". For those dialects class 2a essentially merged into the 2b class. In the northern dialects "eu" merged with "io" which became modern day "ie". The modern Dutch languages has been influenced by both dialects and thus preserves both classes, although the 2a class has considerably shrunk in size.

  • Class 2a verbs: bedriegen, bieden, genieten, gieten, kiezen, klieven, liegen, rieken, schieten, verdrieten, verliezen, vlieden, vliegen, vlieten, vriezen, zieden (rare);
  • Class 2b with historical eu-present: buigen, druipen, kluiven, kruipen, ruiken, schuiven, sluipen, sluiken (rare), sluiten, spuiten, stuiven;
  • Class 2b with historical ū-present: duiken, (ont)luiken, kruien (rare), schuilen, snuiten, snuiven, spruiten, zuigen, zuipen;
  • Historically weak Class 2b roots: fluiten, fuiven (informal), pluizen, wuiven.
  • The verb spugen can also be declined with a class 2 past tense and participle.The verb tijgen has a class 2 past tense and participle when it means 'to pull'.

    Class 3

    Class 3a and 3b have generalised part 3 to part 2, eliminating the -a- from this class.

  • Class 3a roots: beginnen, binden, blinken, (af)dingen, dringen, drinken, dwingen, glimmen, klimmen, klinken, krimpen, ontginnen, schrikken, slinken, spinnen, springen, stinken, verslinden, verzwinden, vinden, winden, winnen, wringen, zingen, zinken, zinnen;
  • Class 3b roots: bergen, delven, gelden, kerven, melken, schelden, smelten, vechten, vlechten, zwelgen, zwellen, zwelten;
  • Two verbs of another class have joined with 3b: treffen (class 5), zwemmen (class 3a)
  • Class 3b verbs that are historically weak: schenden, schenken, trekken, zenden; The verbs became strong by reinterpreting the Rückumlaut that was present in some Old and Middle Dutch weak verbs as a strong vowel alternation.
  • Anomalous: The verb 'worden' (to become) also belonged to class 3b, but the past and present vowels appear to have been swapped: worden werd geworden.
  • Some 3b verbs have a past in -ie- like class 7: helpen hielp geholpen. This can be considered a new "class 3 + 7", and includes (be)derven, helpen, sterven, werpen, werven, zwerven;
  • Class 4

    Class 4 and 5 verbs still show the distinction in vowel between the past singular and plural, although this is not obvious due to the rules of Dutch orthography: ik nam ("I took") has the plural wij namen (not *nammen), that is, the 'short' vowel [ɑ] of the singular is replaced by the 'long' [aː] in the plural. (Note the relationship of consonant doubling to vowel length, which is explained at Dutch orthography). The pattern is therefore: breken brak (braken) gebroken ("to break")

  • Regular Class 4 roots: bevelen, breken, nemen, spreken, steken, stelen;
  • Anomalous: komen the present tense vowel was influenced by a preceding w, which was subsequently lost. The etymological w is retained in the past, unlike English or German: komen kwam kwamen gekomen. The verbs scheren ("to cut"), wegen ("to weigh") and zweren ("to hurt, to sore") have become anomalous. They have replaced the past tense vowel with that of the participle. Thus they decline with a long 'o' in the past tense, like class 2.
  • Class 5

  • Regular class 5 roots: eten, geven, genezen, lezen, meten, treden, vreten, vergeten;
  • Anomalous: zien ("to see") has experienced a loss of the original /h/, with a resulting assimilation of the stem vowel to the vowel of the inflection, and shows Grammatischer Wechsel between this original /h/ and a /g/ in the past: zien zag zagen gezien. The preterite of wezen/zijn ("to be") still shows both (quantitative) ablaut and grammatischer Wechsel between the singular and plural: was/waren.
  • Old j-presents are preserved: bidden, liggen, zitten. These have a short 'i' in part 1 because of the gemination of the consonants, they retain the long 'e' vowel in part 4.
  • Class 6

    Class 6 has become very small, many of its verbs have gone weak or have become semi-strong.

  • Class 6 roots that decline regularly: dragen, graven, varen.
  • Anomalous: slaan (to hit) Like the verb 'zien' has experienced a loss of the original /h/, with a resulting assimilation of the stem vowel to the vowel of the inflection, and shows Grammatischer Wechsel between this original /h/ and a /g/ in the past: slaan - sloeg - sloegen - geslagen
  • The suppleted past tense of the verb 'staan' (to stand) also belonged to this class, it now declines with a short 'o': staan - stond - stonden - gestaan
  • The three inherited j-presents historically decline with 'e'-'oe'-'oe'-'a(a)'. In the modern language they decline irregularly, two have taken 'ie' in the past tense, all three have taken separate vowels in the participle: scheppen schiep geschapen ("to create"), heffen hief geheven ("to lift, raise"), zweren zwoer gezworen ("to swear an oath").
  • This class has the following "semi-strong" verbs with a strong past tense but a weak participle: jagen, klagen (colloquially), vragen, waaien.
  • The verb klagen can be used as a strong verb in colloquial Flemish, but is considered substandard in the modern language.

    Class 7

    Class 7 has shrunk in the modern language, like class 6 many of its verbs have become semi-strong. (The verbs with * are nowadays mostly semi-strong)

  • Class 7a has disappeared. The verb heten ("to call") has become weak, but retains its strong past participle geheten. One relic still remains, but is now declined like a class 1 verb: uitscheiden - scheed uit - uitgescheden ("to secrete")
  • Class 7b: lopen, stoten*
  • Class 7c: houwen, vallen, wassen*
  • One verb displays L-vocalization: houden hield gehouden ("to hold")As in German, two anomalous class 7c verbs have formed new present stems, and shortened the vowel in the past tense: vangen - ving - gevangen ("to catch") and hangen - hing - gehangen ("to hang"). The suppleted past tense of the verb 'gaan' ("to go") also belongs to this class and is declined: gaan - ging - gegaan
  • Class 7d: blazen, slapen, laten, raden*
  • Class 7e: roepen
  • Semi-strong or Mixed verbs

    Many strong verbs have obtained a weak past tense but retain the strong participle malen, moel, gemalen changed into malen, maalde, gemalen. (The verbs with * still use their past tense rarely or archaiacly)

    Class 3: barsten (the older vowels e and o changed to a in standard Dutch, but are still found in some dialects)Class 4: wreken (has retained the long 'o' in the participle)Class 5: wevenClass 6: lachen*, laden, malen, varen ("to fare" The sense "to travel by boat" has a class 6 past voer)Class 7: bakken, bannen, braden, brouwen, heten, raden*, scheiden (uitscheiden has the past scheed uit), spannen, stoten*, vouwen (the past had -ield-, like houden), wassen*, zouten (the past had -ielt-, like houden)A special case is hoeven, which is a weak verb that can decline a strong participle in some circumstances, even though the verb was never strong to begin with.

    Afrikaans

    The distinction between strong and weak verbs has been lost in Afrikaans, as the original past tense has fallen out of use almost entirely, being replaced with the old perfect tense using the past participle. For example, the ancestral Dutch hij zong has become hy het gesing ("he sang/has sung/had sung). One relic of a strong verbs remains, however: wees was gewees ("to be").

    German

    From West Germanic to Old High German:

  • High German consonant shift
  • ai > ei, then ei > ē before r, h and w
  • au > ou, then ou > ō before dentals (þ, d, t, n, l, s, z, r) and h.
  • e > i before u
  • Class 1 has two subclasses, depending on the vowel in the past singular:
  • 1a rītan rītu reit ritum giritan ("to ride")
  • 1b līhan līhu lēh ligum giligan ("to loan" - note grammatischer Wechsel.)
  • Class 2b verbs are rare, unlike in the more northern languages.
  • A few relics of reduplication remain:
  • ana-stōzan ana-sterōz ("to strike")
  • pluozan pleruzzun ("to sacrifice"), in Upper German with the change b > p
  • ki-scrōtan ki-screrōt ("to cut"), in Upper German with the change g > k
  • būan biruun ("to dwell"); this was not a class 7 strong verb originally
  • Changes from Old High German to Modern German:

  • io, ia, ie > ī (spelled <ie>)
  • ou > au
  • ī > ai (spelled <ei>)
  • ū > au
  • ȳ > ɔy (spelled <eu> or <äu>)
  • i > ī (spelled <ie>) before a single consonant.
  • Alternations between past singular and plural are eliminated by generalising part 3 or part 2. If part 3 is generalised in verbs with alternations of the s-r type, it is not just generalised to the past singular but also to the present.
  • The classes are still well preserved in modern German.

  • In class 1, part 3 is generalised, eliminating the older -ei- or -e-. However, a new subdivision arises because the i of the past tense forms is lengthened to ie before a single consonant. reiten ritt geritten ("to ride") versus leihen lieh geliehen ("to loan"). Class 1 verbs in modern German are:
  • with short vowels: beißen, bleichen, gleichen, gleiten, greifen, leiden, pfeifen, reißen, reiten, scheißen, schleichen, schleifen, schleißen, schmeißen, schneiden, schreiten, spleißen, streichen, streiten, weichen (also the originally weak verb kneifen by analogy)
  • with vowel lengthening: bleiben, gedeihen, leihen, meiden, reiben, scheiden, scheinen, schreiben, schreien, schweigen, speien, steigen, treiben, verzeihen, weisen (also the originally weak verb preisen by analogy).
  • In class 2, part 2 is generalised, eliminating older -u-. Class 2b verbs are rare, as in Old High German. They are: biegen, bieten, fliegen, fliehen, fließen, frieren, genießen, gießen, klieben, kriechen, riechen, schieben, schießen, schließen, sprießen, stieben, verlieren, ziehen; with ū-present: saufen, saugen.
  • Two anomalous class 2 verbs in modern German are lügen ("to tell a lie") and trügen ("to deceive"). This no doubt arises from a desire to disambiguate Middle High German liegen from ligen (class 5), which would have sounded the same after vowel lengthening. Trügen would have followed in its wake, because the two words form a common rhyming collocation.
  • In class 3, part 2 is generalised. The o of the 3b participle has been passed by analogy to some 3a verbs, and also to the past of some verbs of both groups: beginnen begann begonnen, bergen barg geborgen ("to rescue"), quellen quoll gequollen ("to well up"). Thus, there are now 5 subgroups:
  • 3a regular (i-a-u): binden, dringen, finden, gelingen, klingen, ringen, schlingen, schwinden, schwingen, singen, sinken, springen, stinken, trinken, zwingen
  • 3a with substitution of o in participle (i-a-o): beginnen, gewinnen, rinnen, schwimmen
  • 3a with substitution of o in preterite and participle (i-o-o): glimmen, klimmen
  • 3b regular (e-a-o): befehlen, bergen; bersten, gelten, helfen, schelten, sterben, verderben, werben, werden, werfen
  • 3b with substitution of o in preterite (e-o-o): dreschen, fechten, flechten, quellen, schmelzen, schwellen
  • In class 4, part 2 is generalised, eliminating older -u- here also. As the only difference between the historical classes 3b and 4 was the preterite plural, these two classes are now identical. Example: nehmen nahm genommen ("to take"). Kommen ("to come") still has the anomalous o in the present stem (although some dialects have regularised it to kemmen): kommen kam gekommen
  • Class 4 verbs in modern German: brechen, gebären, nehmen, schrecken, sprechen, stechen, stehlen, treffen; anomalous: kommen.The preterite of sein ("to be") is Old High German: was/wârum, but levelled in modern German: war/waren.
  • Class 5 is little changed from Old High German. Class 5 verbs in modern German: essen, geben, genesen, geschehen, lesen, messen, sehen, treten, vergessen; with j-presents, bitten, liegen, sitzen. The verb essen ("to eat") had a past participle giezzan in OHG; in MHG this became geezzen which was contracted to gezzen and then re-prefixed to gegezzen.
  • Class 6 is also preserved. However, the j-presents have instead taken an o in the preterite and participle, perhaps by analogy with class 2: heben hob gehoben. The verb schwören has changed e to ö.
  • In Modern German the uo is monophthongised to a u.Class 6 verbs in modern German: fahren, graben, laden, schaffen, schlagen, tragen, waschen; also backen, fragen, though these are usually weak nowadays; with j-present: heben, schwören. The past tense and participle of stehen (stand, older stund, gestanden), which derive from a lost verb *standen, also belong to this class.
  • In class 7, the various past tense vowels have merged into a single uniform -i.e.-. Two verbs have back-formed new present stems from the past stem, and have eliminated grammatischer Wechsel and shortened the vowel in the past tense: fangen fing gefangen ("to catch"), hängen hing gehangen ("to hang")
  • Other class 7 verbs in modern German are: blasen, braten, fallen, halten, heißen, lassen, laufen, raten, rufen, schlafen, stoßen; anomalous: fangen, hängen.The past tense and participle of German gehen, ging gegangen, derive from a lost verb *gangen which belongs to this class. (The verb still exists in other languages, such as the verb gang used in Scotland and northern England.)

    Low German

    The following changes occurred from West Germanic to Old Saxon:

  • ai > ē
  • au > ō
  • eu > io
  • From Old Saxon to Middle Low German:

  • u > o
  • io > e
  • As in Middle Dutch Lengthening of vowels in open syllables: e > ē, o > ō, a > ā, ö > ȫ, ü > ǖ. i Is often lengthened to ē.

    From Middle Low German to Modern Low German:

  • ā > ē
  • ō > ā except before r
  • a > o in preterite forms
  • e > a/ö when followed by two different consonants
  • Most classes are quite well preserved, although the cohesion of some has been lost substantially or even entirely.

  • Class 1 verbs in Low German are bieten, blieven, blieken, diegen/diehen, drieven, glieden, griepen, kieken, lieden, lieken, mieden, rieten, schienen, schieten, schrieden, schrien/schriegen, schrieven, slieken, sliepen, slieten, smieten, snieden, splieten, stiegen, strieden, strieken, swiegen, verdwienen, wieken, wiesen, wrieven and the originally weak verbs glieken, kniepen, priesen by analogy. Some other verbs take either strong or weak past endings: piepen, riesen and spieten.
  • In class 2, part 2 is generalised, eliminating older -u-. Unlike in German but as in Dutch and English, class 2b has grown by moving older class 2a verbs into it. They are beden, bedregen, kesen, legen, flegen, fleten, freren/fresen, geneten, geten, krepen, reken, scheten, spreten, tehn, verleren/verlesen; with ū-present: bugen, krupen, schuven, snuven, sluten, supen, sugen, stuven. The verbs rüken and stöven show anomalous infinitive forms. Some verbs can take either strong or weak past endings: duken and schulen.
  • In class 3, the form of the past participle seems to have been generalised to preterite forms. There are now 5 subgroups + two olders subgroups reduced to one verb each:
  • 3a regular (i-u-u): binnen, dringen, drinken, dwingen, finnen, gelingen, klingen, ringen, slingen, swinnen, swingen, singen, sinken, springen, stinken, wringen. Verbs that may take either strong or weak past endings: blinken, glimmen and klimmen.
  • 3a with ü-infinitive (ü-u-u): begünnen, swümmen
  • 3b regular (a-o-o): bargen, basten/barsten, starven, verdarven, warpen, warrn, warven
  • 3b with ö-infinitive (ö-o-o): hölpen, smölten
  • 3b with e-infinitive and -u- past forms because of phonetical influence of -ll- (e-u-u): gellen, schellen, swellen
  • 3b with e-infinitive (e-o-o): fechten
  • 3b with e-infinitive and different preterite and past participle forms (e-o-a) due to analogy with class 4 verbs: befehlen.
  • In class 4, parts 2 and 3 seem to have merged into -ē-, but due to the influence of past participle forms mostly with a -ō- sound (nowadays written -ā-) a new ending -ō- has arisen:
  • 4 regular (e-o-a): breken, schrecken (with vowel lengthening: schrook, schraken), spreken, stehlen.
  • 4 with two possible preterite forms (e-o/e-a): nehmen, steken
  • 4 with a-infinitive (a-o-a): drapen
  • The verb kamen still shows the -u- infinitive which became -a-: kamen, keem, kamen. The verb to be, wesen, levelled its old preterite forms was/weren into weer/weren, although was still appears in some dialects.
  • In class 5 too the -ē- forms of past participle seem to have influenced the preterite forms. Class 5 regular verbs (ē-ē-ē) include: eten, geven, schehn (preterite scheh or scheeg), lesen (nowadays mostly a weak verb), meten, sehn (preterite seeg) and vergeten. Verbs with j-presents: bidden (sometimes confused with beden), liggen, sitten.
  • The verb treden is anomalous as it has kept the -a- infinitive forms in the preterite and with the variation in vowel length, thus it has tradd, traddst, tradd in the singular with [ɑ] but traden in the plural with [aː]. However, normal class 5 preterite forms treed, treedst, treed, treden may also be found.
  • Class 6 is preserved as well however it has lost its cohesion. Regular class 6 verbs (ā-ō-ā) are graven and slaan (with anomalous infinitive and past participle slaan from earlier slagen). The 3 inherited j-presents have chosen differents paths to make their past forms: heven is now similar to a class 5 verb and has heev in the preterite and heven in the past participle, schapen is a weak verb with strong past participle schapen and swören kept its preterite swoor as well as its past participle sworen - even though it may found with weak past forms.
  • The verb fohren is now merging with föhren and takes weak past endings. The verb dregen has an anomalous infinitive in -ē- but has kept its class 6 past forms droog, drogen (preterite) and dragen (past participle). The verb laden has gone weak but has laden beside laadt in the past participle. The past tense of stahn (stunn), which derives from Middle Low German standen, also belongs to this class.Finally the verb waschen shows preterite wusch and past participle wuschen, just like fallen, fangen and hangen, they seem to make a new strong verb class.
  • In class 7, the various past tense forms have merged into a uniform -ee-.
  • 7a (ē-ē-ē) has one single verb: heten since scheden has gone weak.
  • 7b (ō-ē-ō) also includes one verb: lopen, stoten has gone weak but it kept its strong past participle stoten.
  • 7c has lost cohesion. 7c verb holen (from Old Saxon haldan) has regular heel in the past tense and past participle holen, but fallen, fangen, hangen and gahn (from Old Saxon gangan) show full and fullen, fung and fullen, hung and hungen, gung/güng (but past participlegahn) in the preterite and past participle, all with a short -u-. Class 6 verb waschen has also joined this "new class" and has preterite and past participle wusch and wuschen.
  • 7d (ā-ē-ā) verbs include: laten and slapen, raden and braden are semi-strong as they still have their strong past participles raden and braden (though a weak form braadt may be encountered). Blasen has gone weak.
  • 7e (ō-ē-ō) is reduced to one single verb: ropen. This subgroup had become similar to 7b already in Old Saxon.
  • North Germanic

    Changes from Proto-Germanic to Old Norse:

  • ē > ā
  • a-mutation: u > o when a follows in the next syllable. This affected the past participles of classes 2-4. However, an intervening m or n + consonant blocked this, so the past participle of class 3a kept u.
  • Extension of umlaut to back vowels, causing it to apply also to verbs of class 6.
  • v- is lost before u or o.
  • -n is lost from the infinitive and many inflectional endings.
  • Voiced plosives (but not fricatives) are devoiced word-finally. In Old West Norse, this later causes loss of a preceding nasal.
  • Breaking of e to ja in most environments, and of eu to /.
  • In class 7, several reduplicated verbs are retained: róa reri ("to row"), sá seri ("to sow"), snúa sneri ("to turn").
  • Danish

  • Class 1
  • class 1 strong roots with 'i'supine: blive, drive, fise, glide, gnide, gribe, hive, knibe, lide, pibe, ride, rive, skride, skrige, skrive, slibe, snige, stige, svie, svige, vige, vrideclass 1 strong roots with 'e'supine: bide, fnise, grine, hvine, kige, skide, slide, smide, splide, stride, trine
  • Class 2
  • class 2 strong roots with 'u'supine: bryde, byde, fortryde, *fryse (frøs - frosset/frusset), lyde, skydeclass 2 strong roots with 'y'supine: betyde, flyde, fnyse, gyde, gyse, nyde, nyse, skryde, snydeclass 2 strong roots with 'ø'supine: flyve, fyge, klyve, krybe, kyse, *lyve (løj -løjet), ryge, smyge, syde, stryge
  • Class 3
  • class 3a strong roots with 'u' supine: binde, drikke, finde, rinde, slippe, spinde, springe, svinde, svinge, synge, synke, tvinde, tvinge, vinde, stikke, fornemmeclass 3a strong roots with 'i' supine: *bede (bad - bedt), briste, gide, give, klinge, sidde, stinke, tieclass 3b strong roots with 'u' supine: brække, hjælpe, sprække, trække, træffe,class 3b strong roots with 'æ' supine: hænge, skælve, gælde, knække, kvæde, smælde, smække, være,

    tærske

  • Class 4 has most of its members moved to class 3, only three distinct verbs remain. It is marked by 'a' in the past tense and å in the supine.
  • class 4 strong roots: bære, skære, stjæle,
  • Class 5 this class has also lost cohesion. It is marked by 'å' in the past tense and the supine has the same vowel as the infinitive.
  • class 5 strong roots: ligge, se, æde,anomalous: These two verbs were influenced by a preceding 'w':sove - sov - sovetkomme - kom - kommet
  • Class 6
  • class 6 regular strong roots: drage, fare, jage, lade, tage,anomalous:le - lo - let/leetslå - slog - slået (slaget)stå - stod - ståetsværge - svor - svoret
  • Class 7
  • class 7a: hedde - hed - heddetclass 7b: løbe - løbe - løbetclass 7c:dø - døde - dødfalde - faldt - faldetfå - fik - fåetgå - gik - gåetholde - holdt - holdthugge - hug - huggetclass 7d: græde - græd - grædt

    Norwegian Nynorsk

    Changes from Old Norse to modern Norwegian Nynorsk:

  • á > å
  • Long vowels are usually no longer marked as such: é > e, í > i, ó > o, ú > u, ý > y, œ/ǿ > ø
  • jó/jú > y
  • In class 6, one verb, fara (to fare, travel), has retained its marked long vowel: fór.
  • Multiple of the verbs found in class 7 in Old Norse have gone weak. For instance, although heite (7a) have retained its strong preterite, it has lost its strong supine.
  • Swedish

  • Heta is more commonly inflected weakly: heta, hette, hetat. Similarly, löpa is more commonly weak: löpa, löpte, löpt.
  • References

    Germanic strong verb Wikipedia