7 Verbs

Old English verbs can be daunting, for a typical verb appears in more forms than a typical pronoun, noun or adjective. While no noun has more than six distinct forms, most verbs have fourteen. (Modern English verbs, by contrast, normally have four or five forms.) Further, while some nouns, like mann 'man', have two different vowels in the root syllable, some verbs have as many as five. (The Modern English maximum, leaving aside the verb to be, is three.)

This multiplicity of forms may cause you difficulty when looking up verbs in the dictionary or figuring out their grammatical characteristics. But you can see from the "Magic Sheet" that, despite its inevitable complications, the Old English verb system is really quite orderly. If you keep that orderliness in view as you work through the "Quick Start" section and the rest of this chapter, you will find the verbs to be much easier than they look.

Table 7.1 shows all the forms of two common verbs.

Fremman [By convention, glossaries and dictionaries use the infinitive as the headword for verb entries, and when citing verbs we cite the infinitive.] 'do' belongs to the so-called "weak" class of Old English verbs, those that make the past tense by adding a dental consonant (-d- or -t-) as a suffix. The Old English weak verbs correspond roughly to the Modern English "regular" verbs. Helpan 'help' is a "strong" verb, one that does not add a dental suffix to make its past tense, but rather changes the vowel of its root syllable. The Old English strong verbs correspond to Modern English "irregular" verbs such as sing (past sang, past participle sung).

Take note of these points about the paradigms for fremman and helpan (further details will come later in the chapter):

1. There are just two tenses, past and present. Old English has various strategies for referring to future time: it uses auxiliary verbs (including willan), explicit references to time (e.g. tōmorgen 'tomorrow'), and the simple present, relying on context to express futurity.
2. Similarly, Old English has no settled way of expressing what Modern English expresses with the perfect and pluperfect--that is, that an action is now complete or was complete at some time in the past. It can use forms of the verb habban 'to have' with the past participle, as Modern English does (hæfð onfunden 'has discovered', hæfde onfunden 'had discovered'), it can use the adverb ǣr 'before' with the simple past (ǣr onfand 'had discovered'), or it can use the past tense alone, in which case you must infer the correct translation from the context.
3. While the Modern English verb has only one personal ending (-s for the third-person singular), most Old English verb forms have such endings. These are mostly the same for both weak fremman and strong helpan, but notice that in the singular past indicative the endings are different. The personal endings are shown separately in table 7.2.

   Table 7.2. Personal endings

   		singular	plural
   present indicative	first person	-e	-aþ
   	second person	-st
   	third person	-þ
   past indicative	first person	-e / --	-on
   	second person	-st / -e
   	third person	-e / --
   all subjunctives	all persons	-e	-en

4. Person is distinguished only in the indicative singular, never in the plural or subjunctive. For example, table 7.1 gives the present first-person plural indicative form wē fremmaþ, but the second person is ġē fremmaþ and the third person hīe fremmaþ, with the same verb forms. Further, only the second person is distinguished in the singular past indicative: the first- and third-person forms are the same.
5. The root vowels of strong verbs undergo i-mutation in the present second- and third-person singular indicative: thus the second-person singular of helpan is hilpst, that of faran 'travel' is færst, and that of ċēosan 'choose' is ċīest. The same does not occur in the weak paradigms or in those of strong verbs whose vowels are not subject to i-mutation (e.g. wrītan 'write', second-person singular wrītst).
6. While a Modern English verb descended from the strong verbs never has more than one vowel in the past tense, most Old English strong verbs have two past forms with different vowels, distributed as in table 7.1. The form used for the first- and third-person singular past indicative (e.g. healp) is called the "first past," and the form used everywhere else in the past tense (e.g. hulpon) is called the "second past."
7. The present participle ending in -ende is used where Modern English uses the present participle in -ing: in constructions that express continuing action (for example, "was living") and as adjectives ("the living God").

7.1.2. Preterite-present verbs

Most of the Old English auxiliaries (also called "helping verbs") belong to a class of verbs called "preterite-presents," whose present tenses look like strong past tenses ("preterites") and whose past tenses look like weak pasts. Modern English still has several of these verbs, whose forms we here classify as "present" and "past" according to their origins, even though we no longer think of the past forms as past:

present	past	present	past
can	could	shall	should
may	might	---	ought [The present form of ought is owe, which has, however, become a regular verb with its own past tense.]
---	must [The original present tense of must has been lost.]

The Modern English "present" forms betray their affiliation with the strong past by a peculiarity of their inflection: like a past tense, they lack the ending -s in the third-person singular.

The class of preterite-present verbs is larger in Old English than in Modern English and includes several verbs that are not auxiliaries. Table 7.3 shows the inflection of a few common ones. [Most of the verb paradigms in the remainder of this chapter are abbreviated. You can easily complete them by supplying the correct personal endings.] The present forms of these verbs match the past forms of strong verbs almost precisely: the main differences are that the second-person singular forms have the first past rather than the second past vowel and the ending -st or -t, and the present subjunctive forms sometimes have i-mutation.

The forms in table 7.4 are an amalgam of three different verbs: one that accounts for the present forms in the first column, one that accounts for all the b- forms, and one that accounts for all the w- forms. Paradigms derived from these three verbs overlap, so that there are two complete sets of present forms, [Present forms of the verb wesan (weseð, wesað) are also attested, but they are rare.] two sets of imperatives, two infinitives and two present participles.

The b- forms are often used with reference to future time, as in this sentence on the Day of Judgment:

On þām dæġe ūs bið ætēowed se opena heofon and enġla þrym.

[On that day will be revealed to us the open heaven and the host of angels.]

But the b- forms sometimes are simple presents, as here:

Ðēos wyrt þe man betonican nemneð, hēo biþ cenned on mǣdum and on clǣnum dūnlandum.

[This herb that one calls betony is produced in meadows and in open hilly lands.]

You'll have to look to the context to tell you whether to translate a b- form of bēon as a future.

1. The ending -d- becomes -t- when it immediately follows an unvoiced consonant. The singular past of slǣpan 'sleep' is slǣpte, and that of mētan 'meet' is mētte. The same change occurs in Modern English, though it is not always reflected in the spelling (say reached aloud: what is the final consonant?).
2. The ending -ð becomes -t when it immediately follows d, s or t. For example, the third-person singular of rǣdan 'read' is rǣtt (see also item 3), of rǣsan 'rush' rǣst, and of grētan 'greet' grētt.
3. When a d or g/ġ at the end of a root syllable comes in contact with the ending -st or -ð, it is changed to t or h: for example, the second-person singular of fēdan 'feed' is fētst, and the third-person singular of bīeġan 'bend' is bīehð.
4. Whenever one of these rules has produced a double consonant at the end of a word, or when the ending -ð follows a root ending in ð, the double consonant may be simplified. For example, the third-person singular of ċīdan 'chide' can be ċītt or ċīt, and that of cȳðan 'make known' may be cȳðð or cȳð. A double consonant will always be simplified when preceded by another consonant: so the past singular of sendan 'send' is sende, not *sendde.

Germanic weak verbs fall into three classes: the first two of these are well represented in Old English and the third has almost disappeared (the few remaining class 3 verbs are discussed below). Of the four weak verbs in table 7.5, sceþþan, herian and hǣlan belong to class 1, and lufian belongs to class 2.

Class 1 is marked by i-mutation in the root syllable. If the root is short, gemination (the doubling of the consonant at the end of the root syllable) occurs in certain forms, including the infinitive; but if the consonant is r, you will find -ri- or -rġ- instead of -rr-. The -i- or -ġ- represents a consonant [j], so herian is a two-syllable word: [her-jɑn].

• The geminated form of f is bb (swebban 'put to sleep', third-person singular swefeþ); that of g is cg (bycgan 'buy', third-person singular byġeð).

Class 2 lacks i-mutation. Wherever you find gemination in class 1 verbs with short root syllables, you will find an element spelled -i- or -iġ- after the root syllable of the class 2 verb. [This element did not cause i-mutation because it did not begin with i at the time that i-mutation took place. Rather, it was a long syllable [oːj], which later became the syllable spelled -i-.] This -i- is a syllable all by itself--weighty enough, in fact, to be capable of bearing metrical stress, as we see in this line: [Genesis A, l. 2018. For the metrical notation, see Chapter 13.]

where stress falls on both the first and second syllables of sīðian.

In some verbs, a vowel is inserted before the endings that do not begin with vowels (-st, -ð, -d-). In verbs like sceþþan and herian this vowel is -e-, in verbs like hǣlan the vowel is absent, and in all class 2 weak verbs it is -a- or -o-. Often the vowel is omitted in class 1 verbs with short root syllables, so you can expect to see (for example) fremst and fremþ as well as fremest and fremeþ. This is the rule rather than the exception when the root syllable ends with d or t: so the past tense of āhreddan 'rescue' is āhredde and that of hwettan 'urge' is hwette.

Verbs like Modern English buy/bought, which both change their vowels in the past tense and add the dental consonant characteristic of the weak past, should not be confused with verbs like swim/swam, which are descended from the Old English strong verbs. Buy/bought belongs to a group of class 1 weak verbs in which the vowels of the present tense are subject to i-mutation while the vowels of the past tense are not. Table 7.6 illustrates with cwellan 'kill', sēċan 'seek' and þenċan 'think'.

• The vowels of cwellan are not as predicted in table 2.1 because the unmutated vowel in the forms with e was actually æ, not ea. Similar verbs include cweċċan 'shake' (past cweahte), reċċan 'narrate' (reahte), sellan 'give' (sealde) and tellan 'count, relate' (tealde).

The rule that h is always dropped between vowels (already mentioned in connection with nouns) introduces some complications in the verb paradigm. Table 7.7 illustrates with the class 2 weak verb smēaġan 'ponder'.

Obeying the rule that the most common words are the last to leave a dying class, class 3 contains only habban 'have', libban 'live', secgan 'say' and hycgan 'think' (table 7.8), together with a few odd remnants. Each of these verbs has partly gone over to other classes, and the resulting confusion makes it impractical to describe the characteristics of the class. The best course is to study the paradigms and be prepared to encounter these anomalous verbs in your reading.

Most strong verbs are inflected in pretty much the same way as helpan (table 7.1). You will be able to predict the present paradigm of almost any strong verb if you know how i-mutation affects the vowels of root syllables and how the endings -st and -ð interact with consonants at the ends of root syllables. Once you have learned the gradation patterns for the strong verbs, you will easily master the past paradigms as well.

The Germanic languages have seven classes of strong verbs, each characterized by its own gradation pattern. Gradation is an Indo-European grammatical feature whereby the root vowels of words are altered to signal changes in grammatical function. For example, if the present tense of a Modern English verb contains "short" i followed by n or m, the past-tense form will usually have a and the past participle u: drink, drank, drunk; ring, rang, rung; swim, swam, swum.

Old English has some variations within the Germanic classes, as table 7.9 shows.

This table includes the present third-person singular indicative so that you can see how i-mutation affects each class. You should understand, however, that the vowel of this form is not part of the gradation pattern inherited from Indo-European, but rather a relatively recent phenomenon. Eventually the English language would discard the i-mutation of the second- and third-person singular, but the ancient gradation patterns of the strong verbs are still with us.

• Class 2 verbs like lūcan 'lock' do not conform to the standard gradation pattern; the ū of the present tense has never been satisfactorily explained.
• A few class 3 verbs have u in the present tense. Of these, the one you will meet most frequently is murnan 'mourn' (first past mearn, second past murnon).
• The ō of class 4b appears before nasal consonants. Cuman 'come' belongs to this subclass, but its present tense is anomalous.
• The present tense of class 6 sometimes has æ or ea.
• Class 7 has a variety of vowels in the present tense, not just ā and ō. The past-tense vowels ē and ēo are what distinguish this class.
• You may observe the same gradation patterns you have seen here in families of words derived from the same root. For example, lēof 'beloved' has the same vowel as a class 2 present, ġelēafa 'belief' has the same vowel as the first past, ġelīefan 'believe' has the first past vowel with i-mutation, and lufian 'love' and lof 'praise' have the vowels of the second past and past participle.

Grammatical alternation [A translation of the German phrase "der grammatische Wechsel." In grammars written in English you will usually see it referred to as "Verner's Law" after the Danish linguist Karl Verner, who described its origin. Here we prefer the German term as more descriptive of its function in the recorded language.] is an alternation between one consonant and another to mark the grammar of a word. Only three pairs of consonants alternate in this way:

þ : d     h : g/ġ     s : r

Grammatical alternation affects the paradigms of most strong verbs whose roots end with the consonants þ, h and s: three such verbs are shown in table 7.10.

As you have just seen, some strong verbs are subject to contraction as a result of the loss of h between voiced sounds--the same rule that produces contracted weak verbs. Table 7.11 illustrates with three very common verbs, sēon 'see', slēan 'slay' and fōn 'take'.

• The alternation h/w in sēon is the result of a rare anomaly in the rule of grammatical alternation, the result of which is that ġ varies with w in the second past and past participle. For example, the usual past participle is sewen, but you may sometimes see seġen instead.
• The -n- that appears in some forms of fōn (and also hōn 'hang') was at one time distributed throughout the paradigm. But the rule that n cannot appear before h caused it to be dropped in all forms but those with g. Fōn is also unusual in that the form with g has been extended to the first past (whose vowel is also the same as that of the second past).

A few verbs have the characteristics of the first weak class in the present tense and of strong class 5 or 6 in the past tense. For example, hebban 'lift' has a present tense like that of fremman 'do' or sceþþan 'harm' (tables 7.1, 7.5): iċ hebbe, hē hefeð, etc. But the past third-person singular indicative of this verb is hōf, the plural is hōfon, and the past participle is hafen (the vowel is the same as that of the present, but without i-mutation).

Some common verbs behave in this way, for example, biddan 'ask', licgan 'lie', scieppan 'make, create', sittan 'sit'. The dual nature of these verbs (which most glossaries, including the one in this book, classify as strong) is a curiosity, but it will cause you little difficulty.

In addition to what you learned in 7.1.2, keep these points in mind:

1. The past tense is usually built on the second past root, with -d- or -t- added. In fact, it often looks like the past tense of the weak verbs that change their vowels, though sometimes the forms have been subject to phonological changes that can produce surprising results.
2. When the root syllable ends in g (āgan, dugan, magan), past -d- becomes -t-; g becomes h before that ending and before second-person singular present indicative -t (compare 7.2.1, items 1 and 3).

The present forms of dōn and gān look like those of normal strong verbs. But the past tense of dōn is built on a syllable that looks somewhat like a weak past (though its origin is a mystery), and gān has a past tense that also looks weak and in any case does not belong to the same root that gives us the present forms. Willan looks a bit like a preterite-present verb, but it is not; and its first- and third-person singular present and plural present are quite different from the preterite-present forms.

Most verbs are negated very simply by placing the adverb ne 'not' directly in front of them. In independent clauses, the word-order that follows will normally be Verb--Subject:

Se þe mē ne lufað, ne hylt hē mīne sprǣċe.

[He who does not love me does not keep my sayings.]

Ne is contracted with certain verbs, for example, nis 'is not', næs 'was not' (from bēon), næfð 'does not have', næfde 'did not have' (habban), nyllað 'will not', noldon 'would not' (willan), nāh 'does not have', nāhte 'did not have' (āgan), nāt 'does not know' (witan). Notice that all of the verbs so contracted begin with a vowel, h or w. Not all verbs beginning with those sounds are contracted, but only the more common ones; and those common verbs need not be contracted. You will also see ne wæs, ne hæfð and so on.

The Modern English rule that two negatives make a positive does not apply in Old English; rather, the addition of more negative adverbs to a sentence adds emphasis to its negativity:

Ne bēo ġē nāteshwōn dēade, ðēah ðe ġē of ðām trēowe eton.

[You will certainly not be dead, though you eat from the tree.]

Here the additional negative adverb nāteshwōn makes the sentence more emphatic than it would be with ne alone; since we cannot use double negatives the same way in Modern English, we must resort to a different strategy to represent this emphasis in our translations. Common negative adverbs are nā, nales, nāteshwōn and nātōþæshwōn.

Old English forms periphrastic verbs much as Modern English does, with auxiliary verbs and verbals (infinitives or participles):

auxiliary + infinitive (will find, may find, etc.)
auxiliary + past participle (has found, had found, was found)
to be + present participle (is finding)

This section lists a few ways in which the infinitives and particples of Old English differ from those of Modern English.

Here the subjunctives tell us that the condition described in the noun clause is not a present reality or a future certainty, but a possibility mediated by someone's desire. Some of these usages are disappearing: the first two examples above sound a little archaic, and it would now be more idiomatic to say "The king wanted the knight to go on a quest" and "The king commanded the knight to go on a quest," using infinitive constructions rather than subjunctives.

Aside from these common uses, the subjunctive now appears mainly in fixed or formulaic expressions, for example, "come what may," "thanks be to God."

The subjunctive is far more common in Old English than in Modern English, and you must get used to seeing it in environments where you do not expect it. As in Modern English, the subjunctive is used for conditions contrary to fact. A made-up example:

Ġif iċ wǣre trēowwyrhta . . .

[If I were a carpenter . . .]

It is also used in noun clauses following verbs of desiring and commanding:

Iċ wȳsce þæt iċ wīsra wǣre.

[I wish that I were wiser.]

But the subjunctive is also used in noun clauses where we would not now use it:

Hīe cwǣdon þæt hē wǣre wīs.

Here the subjunctive in the noun clause following Hīe cwǣdon 'They said' does not signal a condition contrary to fact, and cwǣdon 'said' is hardly a verb of desiring or commanding. In fact, the fairest translation of this sentence would be

They said that he was wise.

making no attempt at all to reproduce the subjunctive. What then does the subjunctive express?

Think of it as implying a point of view towards the action of the verb. In clauses following verbs of desire, the point of view is obvious. In Hīe cwǣdon þæt hē wǣre wīs, it is merely that the speaker is reporting an opinion. He is not necessarily taking a position on the rightness or wrongness of that opinion. It may indeed be obvious that he is in complete agreement:

Þæt folc ðā ðe þis tācen ġeseah cwæð þæt Crist wǣre sōð wītega.

[Then the people who saw this sign said that Christ was a true prophet.]

The following sentence is similar, but it uses the indicative:

Be him āwrāt se wītega Isaias þæt hē is stefn clipiendes on wēstene.

[Concerning him (John the Baptist) the prophet Isaiah wrote that he is the voice of one crying in the wilderness.]

The choice between subjunctive and indicative may often be a matter of individual preference or rhetorical emphasis.

Another common environment in which the subjunctive does not necessarily indicate doubt or unreality is the concessive clause introduced by þēah or þēah þe 'though', which generally takes the subjunctive whether or not the statement it contains is known to be true. For example:

Ne sceal nān man swā þēah, þēah hē synful sīe, ġeortrūwian.

[Nevertheless, no man must despair, though he be sinful.]

Here þēah has a sense something like 'even if', implying that the man may or may not be sinful; the subjunctive is appropriate (if a little archaic) even in Modern English. But compare:

God is mildheort, þēah ðe ūre yfelnes him oft ābelge.

[God is merciful, though our wickedness often angers him.]

Here the writer can have no doubt that we do often anger God, but the verb ābelge is still in the subjunctive mood.

In general, you can expect relative clauses, clauses of place, and "when" and "while" clauses to take the indicative. Concessive clauses and "before" and "until" clauses more often take the subjunctive. But the mood in many kinds of clause varies as it does in noun clauses, and linguists argue ceaselessly about the meaning of the subjunctive and the indicative in several common constructions.

The Old English subjunctive is often used to make a first- or third-person imperative, and then the best translation usually converts the subject of the verb into an object of "let." In plural constructions, the -n of the ending is generally dropped.

Sīe hē āmānsumod.

[Let him be excommunicated.]

Ete hīe hrædlīce.

[Let them eat quickly.]

Lufie wē ūre nēxtan.

[Let us love our neighbors.]

This usage survives in some formulaic phrases such as "God be thanked."

Old English Aerobics has a session on the verb, using easy to intermediate Old English sentences. There you will also find a minitext, "Wulfstan's Translation of the Creed," accompanied by exercises on the verbs it contains.