Here belong answere, pt. t. answer-de; make, pt. t. mak-ed, ma-de (for mak-e-de), an extreme example of syncope, pp. mak-ed, maad, mād; clepe, pt. t. clep-ed, clep-te; pley-en, pt. t. pley-de, &c. Also some in which the stem has suffered some alteration, as twicche, pt. t. twigh-te; picche, pt. t. pigh-te; prike, pt. t. prigh-te; reve, pt. t. ref-te, raf-te, pp. raf-t; clothe, pt. t. cladde, cledde, pp. cloth-ed, clad, and even cled; syke, to sigh, pt. s. syk-ed, sigh-te.

Note. The second person of the past tense takes the suffix -st, as in lovedest, contrary to the habit of the strong verbs. An anomalous form occurs in thou made, instead of thou madest.

2. Pt. t. hēr-en, hēr-e, to hear; Pt. s. hĕr-de, Pp. hĕr-d. The vowel is shortened in the pt. s. and pp. before the two consonants. Here belong verbs ending in -an in A.S., which almost invariably exhibit a mutated vowel in the infinitive mood; cf. A.S. sendan, Goth, sandjan.

Here belong: blende, pt. blente; fēde, pt. fedde; fēle, pt. felte; fille, pt. filde; grēte, to greet, pt. grette; hente, pt. hente; hyde, pt. hidde, pp. hid, Kentish hed; kepe, pt. kepte; kisse, pt. kiste, Kentish keste; lede, pt. ledde, ladde; mene, to mean, pt. mente, mēte, to meet, pt. mette; rende, pt. rente; sende, pt. sende, sente; sette, pt. sette; sprēde, pt. spradde; swete, pt. swatte; wende, to go, pt. wente; wēne, to imagine, pt. wende. So also, dēmen, to deem, sēmen, to seem, which should make the pt. tenses demde, semde; but, as these forms seemed awkward, they became demed, semed.

So also lēve, to leave, pt. lefte, lafte; kythe, to make known, pt. kid-de, pp. kid or kythed.

The old combinations enct, engd, became M.E. eynt, eynd. Hence we have blenche, pt. bleynte; drenche, pt. dreynte; quenche, pt. queynte; also the pp. forms y-meynd, seynd, y-spreynd, as if from the infin. menge, senge, sprenge.

3. Infin. tell-en, tell-e; Pt. s. tol-de; Pp. tol-d.

Here tol-de is for an O. Mercian tal-de (A.S. teal-de), from a stem TAL. The infin. shews mutation. The chief key to verbs of this class is to remember that the pt. t. depends upon the original form of the stem, whilst the infin. exhibits mutation; i.e. the pt. t. stem is more original than the present. An old ct becomes ht in A.S., and ght in M.E.

Here belong: leye, also leggen, to lay, pt. layde, leyde; recche, to reck, pt. roghte, roughte; seye, pt. seide, saide; sēke, pt. soghte, soughte; selle, pt. solde; strecche, pt. straughte. Also bye, Kentish begge (in the comp. abegge), to buy, pt. boghte, boughte; werche, to work, pt. wroghte, wroughte (by metathesis for worghte). In a few words a radical n has disappeared before h (M.E. gh) in the past tense: as in bringe, pt. broghte, broughte; thinke, to seem, pt. thoughte (thuuhtə); thenke, to think, pt. thoghte, thoughte (thaohtə, thòuhtə).

Rēche, to reach, tēche, to teach, properly belong to conj. 2; but their past tenses became raughte, taughte, so that they seem to belong here.

The preceding examples give most of the more important weak verbs; others can be found in the Glossary.

Verbs of French origin seldom take -ede in the pt. t., as in the case of aqueyntede; the usual suffix is -ed or -de, or both; as crye, to cry, pt. cry-ed, cry-de; espye, pt. espy-ed, espy-de.

The pp. results from the pt. t. by omitting final -e; if the pt. t. ends in -ed, the pp. coincides with it.

Note. Some verbs have both strong and weak forms; thus abreyde has the str. pt. t. abrayd, and the weak pt. t. abrayde. More striking examples occur in crēpe, to creep, pt. creep, crepte, pp. cropen; slēpe, to sleep, pt. t. sleep and slepte; wepe, to weep, pt. t. weep and wepte. Drede, rede, once strong verbs, are weak in Chaucer; pt. t. dredde, dradde, redde, radde. Cleve, to cleave, has the weak pt. t. clefte, and the strong pp. cloven. Broided is a curious substitution for broiden, the true pp. of breyde (A.S. bregdan). Werien, to wear, is a weak verb of the 1st class; hence the true pt. t. is werede, wered, as in Chaucer. The mod. E. wore is a new formation.

§ 93. Some other verbs. Haven, have, han, to have; pt. t. hadde, also hade; pp. had. A weak verb; often used as auxiliary.

Doon, don, to do. Pres. indic. 1. do, 2. doost, 3. dooth or doth; pl. doon, don. Pres. subj. do; pl. doon, don. Imper. do; pl. dooth, doth. Pp. doon, don. Pt. t. dide (weak). Gerund, to done.

Goon, gon, go, to go. Pres. indic. 1. go, 2. goost or gōst, 3. gooth or gōth, also geeth and gas (Northern); pl. goon, gon, go. Imper. go; pl. gooth. Pp. goon, gon, go; also geen (Northern). The pt. t. is supplied by yede or wente.

Wol, I will. Pres. indic. 1. wol (wil, also written wole); 2. wolt, wilt; 3. wol (also written wole), wil; pl. wollen, woln, wole, wol. Pt. wolde. Pp. wold.

The verb substantive. Infin. been, ben, be. Pres. indic. 1. am^[70], 2. art, 3. is; pl. been, ben, be, beth, rarely aren, are. Pres. subj. be; pl. been, be. Imp. be; pl. beeth, beth. Pp. been, ben, be. Gerund, to bene. Pt. t. 1. was, 2. were, 3. was; pl. weren, were, wer. Pt. t. subj.; were; pl. weren, were.

Anomalous Verbs (Praeterito-praesentia).

Can. Pres. indic. 1. can, 2. canst, 3. can; pl. connen, conne, sometimes can. Pres. subj. conne; pl. connen, conne. Infin. conne. Pt. t. coude, couthe, could, knew. Pp. coud, couth.

Dar. Pres. indic. 1. dar, 2. darst, 3. dar; pl. dar. Pt. t. dorste, durste. Gerund, to durre.

May. Pres. indic. 1. may, 2. mayst, 3. may; pl. mowen, mowe. Pres. subj. mowe, mow. Pt. t. mighte. Infin. mowen.

Moot. Pres. indic. 1. moot (mōt), 2. most, 3. moot (mōt); pl. moten, mote. Pres. subj. mote (but often written moot or mot). Pt. t. moste.

Ow. Pres. indic. 1. ow (?), 2. owest, 3. oweth; pl. owen. Pt. t. oghte, oughte.

Shal. Pres. indic. 1. shal, 2. shalt, 3. shal; pl. shullen, shuln, shul (or shal). Pt. t. sholde, shulde.

Thar. Pres. indic. thar, impersonal. Pt. t. thurfte, hurte, impersonal.

Woot. Pres. indic. 1. wòòt (wot), 2. wòòst (wost), 3. wòòt (wot); pl. witen, wite, also woot (incorrectly). Pres. subj. wite. Infin. witen, wite; also weten. Pt. t. wiste. Pp. wist.

§ 94. Negative forms. Ne, not, is prefixed to some verbal forms, and coalesces with them.

Ex. nam, for ne am; nart, for ne art; nis, for ne is; nas, for ne was; nere, for ne were. Nadde, ne hadde; nadstow, ne haddest thou; nath, ne hath. Nil, ne wil; niltow, ne wilt thou; nolde, ne wolde. Noot, ne woot; niste, ne wiste. We even find nacheveth written for ne acheveth; &c. Cf. nof, for ne of; nin for ne in.

§ 95. Adverbs.

Some adverbs are formed by adding -e to the adjectival form; as dēp-e, deeply, from deep, A 129; loud-e, loudly, from loud, A 714. Hence, beside the usual forms heer, here, ther, there, wher, where, eek, eke, we find the anomalous forms her-e, ther-e, wher-e, ek-e; which we should hardly expect. So also moste, E 1714, F 1622, as well as most; probably because the word the precedes, which suggested the definite adjectival form, though the word is really used adverbially. Other double forms are thanne, than, then; whanne, whan, when. Amongst other forms in -e may be mentioned: asyde, atwinne, bihinde, bisyde, bothe, nouthe, ofte, selde, sone. Remarkable forms are ther-fore, wher-fore (see Stratmann). Some forms result from loss of n, as aboute from abouten; so also above, bifore (also biforn), henne, inne, withoute; cf. binethen, sithen.

Many adverbs are characterised by the suffix -es; as agates, amiddes, amonges, bisydes, bitymes, elles, nedes, togidres, unnethes. So also hennes, thennes, whennes; ones, twyes, thryes. The gen. suffix -es appears clearly in his thankes, A 1626.

Some adverbs have an internal -e-, which is not found in A.S., as in bold-e-ly, A.S. bealdlīce; and this -e- counts as a syllable. So also nedely, D 968 (but nēd(e)lý in B 4434); softely, E 323; trewely, A 773. So also semely, rudeliche.

Other noteworthy adverbs are: bet, better; fer, far, comparative ferre; negh, nigh, neer, ner, nearer; leng, lenger, longer; mo, more; more, more; uppe, up.

§ 96. Prepositions and Conjunctions.

These are given in the Glossary. We may note the occasional use of the form til (usually Northern) for to, chiefly before a vowel. Also the use of ne ... ne for neither ... nor; other ... other, either ... or; what ... what, partly ... partly; what for ... and, both for ... and; what with ... and, both by ... and.

§ 97. Constructions. Amongst unusual constructions we may particularly note the position of with, when used adverbially. In such a case, it is immediately subjoined to the verb, instead of being separated from it as in mod. E. Ex. 'to shorte with your weye,' to shorten your way with, A 791; 'to helen with this hauk,' to heal this hawk with, F 641.

Another remarkable construction is seen in such a phrase as 'The kinges meting Pharao,' the dream of king Pharaoh; see note to F 209.

At the beginning of a sentence ther frequently means 'where'; it makes all the difference to the sense.

§ 98. Versification.

The structure of English versification has been much obscured by the use of classical terms in senses for which they are ill-adapted, and by artificial and wooden systems of prosody which obscure the natural pronunciation of sentences. In order to prevent all obscurity, the terms employed shall be carefully defined.

Strong and weak syllables. An accented syllable is strong, An unaccented syllable is weak. A syllable that bears a secondary or a slight emphasis is half-strong. A very weak or slightly pronounced syllable is light.

Examples. In the words light, alight, lighter, the syllable light is, in each case, 'strong'; the syllables a- and -er are 'weak.' Chaucer sometimes uses such a word as light-e, in which the final -e may constitute a syllable of the verse, in which case it is 'weak'; or it may be elided or nearly elided before a vowel, in which case it may conveniently be described as being 'light.' In such a word as cónqueròr, there are really two accents. The true 'strong' accent is now on the first syllable; the 'half-strong' or secondary accent is on the third syllable; and it is not unusual to denote this by the use of an acute accent for the strong, and grave accent for the half-strong syllable.

§ 99. Three Latin terms. A word such as alight is often described as constituting an 'iambus' or 'iamb'; and I shall sometimes here use this term, but under protest. An iambus is properly a short syllable followed by a long one; whereas the English iamb is a weak syllable followed by a strong one, which is a very different thing. The confusion between length in Latin verses and strength in English verses is pernicious, and has greatly misled many writers on metre; for the difference between them is fundamental.

In the same way, such a word as lighter may be called a 'trochee'; but it must never be forgotten that, in English poetry, it means a strong syllable followed by a weak one, and is independent of the notion of 'length.'

Similarly, such a word as alighted, in which a strong syllable is situated between two weak ones, may be called an 'amphibrach.' The amphibrach plays a highly important part in English verse, though it is usual not to mention it at all. I shall use these three terms, iamb, trochee, and amphibrach, only occasionally, and for the convenience of the names; it being now well understood that I merely mean such groups of strong and weak syllables as occur in the English words alight, lighter, and alighted.

Having thus explained that an 'iamb' has nothing to do with long and short syllables, I shall nevertheless use, to denote it, the ordinary symbol weak strong. Similarly, the symbol strong weak means a trochee; and the symbol weak strong weak means an amphibrach. It follows that weak here means, not a short, but a weak syllable; and strong here means, not a long, but a strong one. If this be remembered, all will be clear; but not otherwise.

§ 100. I shall attempt, first, to describe the versification of the lines in the Canterbury Tales; it will be easy to explain the shorter lines (of four accents) afterwards.

Speech-waves. In English, accent plays a very important part; and for this reason, we may consider English speech as consisting of a succession of utterances which form, as it were, speech-waves, in which each wave or jet of breath contains a strong syllable; and this strong syllable may either stand alone, or may be preceded or followed by a weak syllable, or may even be both preceded and followed by a weak syllable during the emission of the same jet of breath^[71].

Thus each jet of breath, due to a slight impulse emitting inhaled air, may be denoted by strong, or by weak strong, or by strong weak, or by weak strong weak. That is, the words light, alight, lighted, alighted can all be produced in a single speech-wave. But if a word has two accents, it requires two impulses to utter it, and really contains two speech-waves. Such words are extremely common; as cónque-ròr, amál-gamàte, &c.; and many English words require three speech-waves, as insòl-ubíli-tỳ; or even four, as ìn-combùsti-bíli-tỳ.

§ 101. Here comes in the distinction between prose and verse. It is equally easy to describe the accentual structure of either; and it is readily perceived that, in prose, the speech-waves succeed each other so that there is, usually, no perceptible regularity in the distribution of strong and weak syllables; but, in verse, we expect them to be distributed in a manner sufficiently regular for the ear to recognise some law of recurrence, and to expect it.

An extremely regular line occurs in Goldsmith's Deserted Village:—

This obviously consists of five consecutive iambs, and may be denoted by: weak strong . weak strong . weak strong . weak strong . weak strong. Here the dot (.) is introduced to shew precisely where the natural pause in the voice, or the separation of the speech-waves, occurs.

It is usual, in books of prosody, to introduce a bar instead of a dot, and thus to break up the line into bits of equal length, and to exhibit the result as the Procrustean formula to which all lines of five accents should be reduced. There is little to be learnt from this wooden method, which amounts to little more than leaving the reader to find out the scansion for himself as he best may; for few lines really conform to it.

If, bidding adieu to this artificial system, we inquire into the way in which a good reader really articulates the lines, we find that he, following the poet, is so far from conforming to this uniform type of line, that he usually does his best to avoid it; and the more skilfully he does this, the more he is appreciated for his variety. Indeed, the number of possible variations is considerable, as Goldsmith may again teach us, if, instead of using a bar to denote the artificial pause, we use a dot to denote the natural and the actual one. Good examples occur in the following lines, all different in their effect. Observe that the hyphen is used to bring together words that are pronounced in a single speech-wave; for just as cónque.rór requires two jets of breath, it often happens that two words (one of them enclitic) can be uttered in one.

These may be analysed as below.

These three lines are obviously different, and all differ from the line already quoted.

If, however, we now remove the dots, all four lines can be included in the same formula: weak strong weak strong weak strong weak strong weak strong. And this is what is really meant (or ought to be meant) by saying that Goldsmith's line consists of five iambic feet; the general type weak strong being called an iambic foot.

§ 102. As the use of dots, as above, is rather confusing, we might employ the usual bars instead; assigning to them natural instead of artificial positions. But it will be better, under the circumstances, to employ special types. I shall use strong to denote a strong syllable, and half strong to denote a half-strong syllable. Then, if the weak syllable be denoted by a thin up-stroke or down-stroke, we have iamb to denote an iamb; trochee for a trochee; and amphibrach for an amphibrach; and the four lines from Goldsmith may be thus scanned^[72]:—

In every case an upstroke is followed by a horizontal one, i.e. a weak syllable by a strong one, but the general effect is variable, and is easily caught by the eye. This method at once detects a real recurrence of a line cast in precisely the same mould. Thus the line—'For-talking age and-whispering lovers made' is to be scanned: amphibrach strong amphibrach trochee strong and thus closely resembles the third of the above lines, being denoted by the same formula.

§ 103. When we come to apply a similar system of scansion to Chaucer, we find that he differs from Goldsmith in FOUR important particulars. This is because he followed, more immediately, the rules of verse as exhibited in the Old French metres. I quote the following from P. Toynbee's Specimens of Old French, p. liii:—

'In ten-syllabled lines [i.e. in lines of five accents] the pause or caesura is after the fourth syllable:—

At the caesura, and also at the end of the line, a feminine syllable [i.e. a weak or light additional syllable] is admissible, which does not count, even if it is not elided. It is thus possible to have no less than four different forms of ten-syllabled epic lines, all equally correct; viz.

Here, in (b) and (d), there is an additional syllable at the caesura or middle pause; and, in (c) and (d) there is an additional syllable at the end of the line. Hence the number of syllables is, in (a), ten; in (b) and (c), eleven; and in (d) twelve. But the number of accents is the same in all, viz. five. It is therefore better to speak of these lines as containing five accents than to call them ten-syllabled lines.

All the above varieties are found in Chaucer; and we thus see TWO of the particulars in which he differs from Goldsmith, viz. (1) that he sometimes introduces an additional syllable at the end of the line; and (2) that he does the same after the caesura, or at what may (roughly) be called the end of the half-line.

§ 104. But the fact is that Old French verse admits of more licences than the above. It was also permissible for the poet (besides adding to the line at the end) to subtract from it at the beginning, viz. by omitting the first weak syllable at the beginning, or the first weak syllable in the second half-line; i.e. after the caesura. This accounts for TWO MORE particulars of variation from the modern line of Goldsmith.

The result is that the Old French verse absolutely exhibited no less than sixteen varieties; and the actual number of syllables varied from eight (the least) to twelve (the greatest number). Dr. Schipper gives the true scheme in his Englische Metrik, p. 440, as follows; where the number following each scheme expresses the number of syllables.

§ 105. Thus Chaucer had, unquestionably, sixteen forms of verse to choose from. It only remains to discover how many of these he actually employed.

The shortest answer is, that he freely accepted the principles of adding a syllable at the end of the line and at the end of the half-line. He also allowed himself to accept the principle of dropping the first syllable of the line^[73]. But he disliked forms 9, 11, 13, and 15, which introduce a most disagreeable jerk into the middle of the line, such as he very rarely allows^[74].

§ 106. The general rules for the mode of reading Chaucer's lines have been given above (§ 67); and need not be here repeated.

I now subjoin some examples. In each case the prefixed number refers to one of the sixteen forms given in § 104; whilst the symbols following the lines give the natural method of scansion. Words joined by hyphens are pronounced in the same jet of breath. I may also note here that a trochee is sometimes substituted for an iamb, i.e. trochee for iamb; especially at the beginning of a line, or of the latter half-line. The place of the caesura is denoted by a bar. A shorter down-stroke than usual signifies a light syllable, as defined in § 98. The following examples are from Group A of the Canterbury Tales:—