Top of chapter Contents Bibliography 15.1. Inflections
15.1.1. Pronouns 15.1.2. Verbs 15.1.3. Adjectives
15.2. Syntax
15.2.1. Omission of subjects and objects 15.2.2. Omission of prepositions 15.2.3. Adjectives used as nouns 15.2.4. Word-order 15.2.5. Independent and subordinate clauses

15. The Grammar of Poetry

You are already aware of some of the grammatical differences between prose and poetry. You know, for example, that Old English poetry has some rules of its own for the ordering of sentence elements, and you know that poetry makes heavy use of apposition. Here we will discuss the grammar of poetry in greater detail.

15.1. Inflections

15.1.1. Pronouns

You will frequently see accusative singular þec ‘you’ and mec ‘me’ where prose has þē and .

Instead of the genitive singular pronoun his, you will sometimes see sīn ‘his’ used as a possessive adjective. It takes strong adjective endings.

15.1.2. Verbs

You may (rarely) see a present first-person singular verb with the archaic ending -o or -u: for example fullǣstu ‘assist’ in Beowulf, l. 2668, but more often hafo, hafu ‘I have’ instead of West Saxon habbe.

The present second-person singular and third-person singular endings are -st and in West Saxon (see table 7.2.). But in poetry, which frequently displays northern dialect features, you will often see -est and -eð instead. And where West Saxon has i-mutation of the root vowel, these longer forms generally lack it. For example, the West Saxon present third-person singular of healdan ‘hold’ is hielt, but you will see healdeð in poetry; and the West Saxon present third-person singular of brūcan ‘make use of’ is brȳcð, but you will see brūceð in poetry.

Certain archaic and dialectal verb forms occur in both prose and poetry, but more often in poetry. These include cwōm (past tense of cuman ‘come’), sǣgon, sēgon (past plural of sēon ‘see’), ġēong (past tense of gangan ‘to go’), and alternate forms of third-class weak verbs, especially hafast, hafað beside hæfst, hæfð (forms of habban ‘have’).

15.1.3. Adjectives

In poetry, weak adjectives are frequently found where you would normally find strong adjectives in prose—that is, where no demonstrative pronoun or possessive adjective precedes (for the usual rule, see the beginning of Chapter 8). Example:

wolde blondenfeax   beddes nēosan
gamela Scylding.
[the gray-haired one, the old Scylding, wished
to seek his bed.]   (Beowulf, ll. 1791-92)

The strong form corresponding to gamela ‘old’ would be gamol.

The reverse does not happen: strong adjectives are not used with preceding pronouns or possessive adjectives. You will never see such phrases as *þone gōdne cyning.

15.2. Syntax

15.2.1. Omission of subjects and objects

You learned in an earlier chapter that a pronoun subject may be omitted in Old English. In fact, when reading poetry you will frequently encounter clauses with unexpressed subjects. Often it is no more than a matter of one subject belonging with two predicates:

Ðā ārās mæniġ goldhladen ðeġn,   ġyrde hine his swurde.
[Then many a gold-laden thegn arose (and) girded his sword on himself.]
(The Battle of Finnsburh, l. 13)

As the translation suggests, we can do much the same thing in Modern English, though we usually say and between the two predicates. But sometimes it is not so easy to figure out the reference of an unexpressed subject:

Sceolde lǣndaga
æþeling ǣrgōd   ende ġebīdan,
worulde līfes,   ond se wyrm somod,
þēah ðe hordwelan   hēolde lange.
[The good old prince
had to experience the end of his transitory days,
of his life in the world, and the worm along with him,
though (he) had held the hoard-wealth for a long time.]
(Beowulf, ll. 2341-44)

The subject of the clause in the last line is evidently the dragon (which has been guarding the only treasure that interests us in the last third of Beowulf), but the subject of the preceding clause, being compound, does not match it precisely.

In the examples above, the reference of the unexpressed subject is someone or something that has recently been mentioned. But the unexpressed subject need not have an antecedent:

Þǣr mæġ nihta ġehwǣm   nīðwundor sēon
fȳr on flōde.
[There every night (you) may see an evil wonder,
fire in the water.]   (Beowulf, ll. 1365-66)

Here it is a simple matter to supply a pronoun subject.

In Modern English we often use it as a subject when there is no logical need for a subject: in the phrase “it rained,” we would be hard put to identify the “it” that is doing the raining. Old English uses the same idiom, but it does not have to; and you will find that poetry often leaves the subject unexpressed where we would say “it”:

Nāp nihtscūa,   norþan snīwde
[The night-shadow darkened, (it) snowed from the north]
(The Seafarer, l. 31)

Direct objects may also be omitted. Usually the object will be expressed in a nearby clause (though not always as an object):

Ðā ġȳt hīe him āsetton   seġen gyldenne
hēah ofer hēafod,   lēton holm beran,
ġēafon on gārsecg.
[Then further they set up for him a golden standard,
high over head, let the sea bear (him),
gave (him) unto the sea.]   (Beowulf, ll. 47-49)

There can be no doubt as to whom they are sending out onto the sea; it is the one for whom they set up the standard.

15.2.2. Omission of prepositions

You will remember from Chapter 4 that words in the dative case are often used by themselves where Modern English would use a preposition. This tendency is even more pronounced in poetry than in prose. Examples:

Weorða ðē selfne
gōdum dǣdum   ðenden ðīn God reċċe.
[Honor yourself
with good deeds for as long as God cares for you.]
(Waldere, I, ll. 22-23)
þonne hand wereð
feorhhord fēondum.
[when my hand defends
my life-hoard from enemies.]   (Waldere, II, ll. 21-22)
seġe þīnum lēodum   miċċle lāþre spell,
[say to your people a much more hateful message]
(The Battle of Maldon, l. 50)

As you can see, you will frequently have to supply a preposition when you encounter a word in the dative that lacks one. But there is no one Modern English preposition that is always appropriate. You will have to judge from the context what the dative is doing and how best to translate it.

In the first passage above, notice also the clause ðenden ðīn God reċċe ‘for as long as God cares for you’. Here the verb reċċan takes the genitive of what one cares for, and we supplied a preposition in translating it. Verbs that govern words in the genitive case are common in both verse and prose. For example, ġielpan ‘boast’ takes the genitive of what one is boasting of (you must supply the preposition of or about) and þancian ‘thank’ takes the genitive of what one is grateful for (you must supply the preposition for). A good glossary or dictionary will tell you about the cases that verbs govern.

15.2.3. Adjectives used as nouns

In Modern English, when we wish to name a thing by mentioning one of its attributes, we use an adjective with a placeholder noun: “the wise one,” “the big one.” In Old English poetry, it is more common to use a demonstrative pronoun with a weak adjective:

Þā wæs Nerġendes
þēowen þrymful,   þearle ġemyndiġ
hū hēo þone atolan   ēaðost mihte
ealdre benǣman   ǣr se unsȳfra,
womfull onwōce.
[Then the Savior’s handmaiden
was filled with glory, vigorously thoughtful
how she could most easily deprive
the terrible one of life before the unclean one,
the impure one awoke.]   (Judith, ll. 73-77)

Here Holofernus (about to be beheaded by Judith) is þone atolan ‘the terrible one’, se unsȳfra ‘the unclean one’, and finally womfull ‘the impure one’. The last of these is a strong adjective unaccompanied by either a demonstrative or a noun. Strong adjectives are used as nouns less often than weak adjectives are, but it happens often enough that you should be prepared for it.

15.2.4. Word-order

The basic patterns of Old English word-order that you learned in Chapter 12 apply as well for poetry as they do for prose. To illustrate, here is a short passage with the word-order of each clause indicated:

1. Verb-Subject:
Ðā wearð breahtm hæfen.
2. Verb-Subject:
Beorg ymbstōdan
hwearfum wræcmæcgas.
3. Subject-Verb:
Wōð ūp āstāg
ċearfulra ċirm.
4. Verb-Subject:
Cleopedon moniġe
fēonda foresprecan,   firenum gulpon:
5. Subject-Verb:
“Oft wē ofersēgon   bi sǣm twēonum
þēoda þēawas,   þræce mōdiġra
6. Subject . . . Verb
þāra þe in ġelimpe   līfe wēoldon.”
[1. Then a cry was raised. 2. The devils stood
around the mound in crowds. 3. The noise, the uproar
of the miserable ones rose up. 4. Many advocates
for the enemies called out, boasted criminally:
5. “Often we have observed, between the two seas,
the customs of the nations, the power of those proud ones
6. who lived their lives in prosperity.”]
(Guthlac, ll. 262-68)

Each clause in this passage (chosen nearly at random) uses a standard word order. If the passage seems difficult, that is because the poet is vigorously taking advantage of the flexibility of these standard word orders. For example, in 1. the finite verb is an auxiliary, and the verbal (a past participle) is delayed to the end of the clause, and in 2. the direct object comes before the verb instead of after the subject).

Variation or, to use the grammatical term, apposition, would seem likely to violate the norms of Old English word order. In 3., the subject Wōð is varied by ċearfulra ċirm, and thus a subject follows as well as precedes the verb: the word-order is really Subject-Verb-Subject. But you will often find that it is possible to look at such clauses as hybrids of two standard word-orders: in this case Subject-Verb and Verb-Subject. Clause 4., where the word-order is Verb-Subject-Verb, can also be seen as a hybrid. It is as if poets saw the clause as containing several positions where a subject, verb or other element would be permissible and set out to fill up those positions.

It would be nice if you could always count on elements in variation coming in “normal” positions, but sometimes they do not:

Hē ǣrest sceōp   eorðan bearnum
heofon tō hrōfe,   hāliġ Scyppend.
[he, the holy Creator, first created
heaven as a roof for the children of men.]
(Cædmon’s Hymn, ll. 5-6)

The beginning of this sentence, with its order Subject-Verb . . . Object, looks normal enough, but the variation hāliġ Scyppend comes where a subject normally does not come (as part of a sequence Verb . . . Object-Subject). This example should serve as a reminder that you must be especially attentive to grammatical form and context when reading poetry. We can tell that hāliġ Scyppend is a subject, in variation with , because it is nominative in form and because the poem has been talking about God.

15.2.5. Independent and subordinate clauses

In §10.2-10.4 you learned that some adverbs have the same form as conjunctions and that the two occur together in correlative constructions. In §12.5 you learned further that word-order will often tell you which clause of a correlative construction is independent and which is subordinate. We also warned you there, however, that the word-order rule does not work in poetry. So how can you tell, in a sentence like the one that follows, whether we have a correlative construction, and if we do, which clause is independent? (We omit editorial punctuation to discourage you from prejudging the case.)

Ðā wæs on ūhtan   mid ǣrdæġe
Grendles gūðcræft   gumum undyrne
þā wæs æfter wiste   wōp up āhafen
miċel morgenswēġ
[When/Then Grendel’s warcraft was manifest
to men at dawn, early in the day,
when/then after the feasting weeping, a great morning-sound,
was raised up.]   (Beowulf, ll. 126-29)

Even where we don’t have ambiguous adverb/conjunction pairs, it can be difficult to distinguish independent and subordinate clauses:

Nū ēow is ġerȳmed   gāð ricene tō ūs
guman tō gūþe   (The Battle of Maldon, ll. 93-94)

If is an adverb, the translation should go like this:

Now the way is open to you; go quickly to us,
men to battle.

But if is a conjunction, it should go like this instead:

Now that the way is open to you, go quickly to us,
men to battle.

How to read such sentences as these is a matter of controversy. Until around the middle of the twentieth century, editors more often than not interpreted ambiguous clauses as independent and supplied punctuation to match that interpretation. In any case, editors showed an aversion to sentences in which subordinate clauses preceded independent clauses. In a passage like the following, we have a choice of translating Þā as “then” and punctuating the first clause with a semicolon or translating Þā as “when” and punctuating with a comma:

Þā of wealle ġeseah   weard Scildinga
se þe holmclifu   healdan scolde
beran ofer bolcan   beorhte randas
fyrdsearu fūslicu   hine fyrwyt bræc
mōdġehygdum   hwæt þā men wǣron.
[Then/when the guardian of the Scyldings, he who
had to hold the sea-cliffs, saw from the wall
(them) bearing their bright shields, their ready army-trappings,
over the gangway;/, curiosity tormented him
in his mind-thoughts (to know) what those men were.]
(Beowulf, ll. 229-33)

Early editors and translators would almost invariably choose “then” and the semicolon. Recent editors are more likely to interpret the first clause as subordinate and punctuate with a comma.

Our decision whether to interpret a clause as independent or subordinate rarely makes much difference in the sense of a passage, but it does make a significant difference in the way we perceive its style. A paratactic style (one with relatively few subordinate clauses) was once thought to be “primitive,” especially by scholars who were interested in recovering, in Old English poetry, a genuine experience of English or Germanic cultural origins. Now, on the other hand, scholars are more likely to deny the possibility (and perhaps also the value) of recovering the origins of a culture, and further to deny that parataxis is in any way “primitive.” Such modern scholars have been open to arguments that Old English poetry is less paratactic than formerly believed.

But how can you decide, in a particular passage, whether a clause is independent or subordinate? The following rule seems to work for clauses that contain an auxiliary and a verbal: if the auxiliary precedes the verbal and is unstressed, the clause is independent, but if the auxiliary follows the verbal and is stressed, the clause is subordinate. So this clause, in which the auxiliary wearð precedes the verbal ġeġearewod and is unstressed, is independent:

Þā wearð snelra werod   snūde ġeġearewod,
cēnra tō campe.
[Then the host of the bold and the brave was quickly
prepared for battle.]   (Judith, ll. 199-200)

This clause, on the other hand, in which a stressed auxiliary (hafað) follows the verbal (ġetācnod), is subordinate:

swā ēow ġetācnod hafað
mihtiġ Dryhten   þurh mīne hand.
[as the mighty Lord
has signalled to you through my hand.]
(Judith, ll. 197-98)

It may be uncertain whether clauses in which stressed auxiliaries precede verbals, or which do not contain auxiliaries, are independent or subordinate—unless, of course, the context tells us, as it often does.

The existence of clauses that may be either independent or subordinate has occasioned debate, some holding that Old English had a type of clause that fell somewhere between independent and subordinate while others believe that Old English clauses were always one or the other, even if we do not always know how to distinguish them. In this connection it is worth noting that the rule for distinguishing independent and subordinate clauses that contain auxiliaries was not discovered until relatively recently (see Donoghue 1987). It is not inconceivable that a rule for distinguishing other clauses has yet to be discovered.