Chapter XIV · —THE CONSTRUCTION OF SENTENCES, IDIOMS, ETC.

§ 1. In later Cornish there was a strong tendency to assimilate the order of words and the construction of sentences to those of English, but nevertheless certain idioms persisted throughout.

In English the normal order of words in a simple sentence is:—

Subject—Verb—Complement of Predicate (Object, etc.).

This order is used in Cornish also when the impersonal form of the main verb or of the auxiliary is used, and the object is not a personal pronoun. Thus:—

Dew a gar an bês, God loveth the world.

Dew a wra cara an bês, God doth love the world.

One of these two forms is the most usual in a direct affirmative principal sentence when the object is not a pronoun.

If the object is a pronoun, the order is:—

Subject—Particle—Object—Verb.

Thus:—

Dew a’th gar, God loveth thee.

Or:—

Subject—Particle—Auxiliary—Pronoun in the Possessive Form—Infinitive of Main Verb.

Thus:—

Dew a wra dha gara, God doth love thee.

If the auxiliary verb is bos, to be, it often happens that the inflected form of it is used in an affirmative sentence when the tense is the continuous present or imperfect. In these cases the order is:—

Auxiliary Verb—Subject—Participle of Main Verb—Complement.

Thus:—

Thov vî ow môs dhô Loundres, I am going to London.

Therough why ow tôs adre, you were coming home.

But with the preterite tense the simple impersonal form is more usual. Thus:—

Mî a vê gennes en Kernow, I was born in Cornwall.

The same applies to the present and imperfect of bos when it is not an auxiliary. Thus:—

Thov vî lowen dhô ’gas gwelas, I am glad to see you.

The inflected form of the verb is rare in simple direct affirmative sentences, except when it is used as a Celtic substitute for “yes.” It may be used in verse, but it is rather affected in prose. In negative, interrogative, and dependent sentences it is the only form to use, but even then it is the inflected auxiliaries, parts of gwîl, to do, menny, to will, gally, to be able, etc., with the infinitive of the main verb that are more commonly used, rather than the inflected form of the main verb itself. In the third person singular it is of course only distinguishable from the impersonal form by the position of the subject, which in the inflected form would follow the verb. The inflectional form of the third person plural is only used when the pronoun “they” is the subject. When the subject is a plural noun the verb is always in the singular. The inflected form, either of the auxiliary gwîl with the infinitive of the main verb, or of the main verb itself, is always used for the imperative. In late Cornish, except in the case of answers for “yes” and “no,” and of the peculiar forms of the first, second, and third persons singular in ma, ta, and va, the subject personal pronoun is almost always expressed, except, of course, when the subject is a noun.

§ 2. Negative and Interrogative Sentences.

For a negative sentence it is never correct to use the impersonal, but always the inflected form of the verb or of the auxiliary, preceded by the negative particle ni (older, ny) or nyns. The order is:—

1. Negative Particle—Verb—Subject—Complement of Predicate.

Or:—

2. Negative Particle—Auxiliary (inflected)—Subject—Infinitive of Main Verb—Complement.

If the object is a pronoun, in the first case it follows the negative particle in its third form, in the second case it precedes the infinitive in the possessive form. Thus:—

1. Ni welav vî an dên, I do not see the man.

Ni wôr dên vîth an êr, no man knoweth the hour.

2. Ni wrígav vî gwelas an dên, I did not see the man.

1. Ni’th welav vî, I do not see thee.

2. Ni wrigav vî dha welas, I did not see thee.

In the case of the present and imperfect of bos, to be, the particle nyns is often used, and it is sometimes found with other words beginning with vowels, but its use is rare in late Cornish, and ni, or less correctly na (or nag before a vowel) is more usual.

It is allowable to use the inflected form with the subject-pronoun preceding the negative particle, but it should only be used for emphasis on the subject, and is better avoided.

Interrogative sentences are formed with the interrogative particle a, or by the use of some interrogative pronoun or adverb. In all cases the inflected form of the main verb or auxiliary (usually the latter) follows the particle, pronoun, or adverb, and usually with its initial in the second state. Thus:—

A wrîgough why besca gwelas? Did you ever see?

A wreugh why agan gwelas? Do you see us?

Fatla wreugh why crŷa hedna? How do you call that?

Fraga wreugh why gwîl hebma? Why do you do this?

A vednough why môs genev vî? Will you go with me?

The particle a is often omitted colloquially, but its effect is perceptible in the change of the initial of the verb. If the verb begins with a vowel, a is always omitted.

With interrogative sentences should come the answers to them. It must be understood that by nature no Celt can ever say a plain “yes” or “no.” There are “dictionary words” for “yes” and “no” in Welsh and Cornish, and they are used a very little in translations from other languages; but they do not “belong” to be used in speaking or writing Welsh or Cornish. In Gaelic there are not even “dictionary words” for them. In Breton ia and nan are used freely for “yes” and “no,” as in French, but that is probably quite modern French influence. The Celtic practice is to repeat the inflected verb of the question, affirmatively or negatively, in the necessary person. Thus:—

’Ellough why cowsa Kernûak? Can you speak Cornish?

Gellam or mî ellam. I can (yes). Ni ellam, or (less correctly) nag ellam, I cannot (no).

A vednough why dôs genev vî? Will you come with me?

Mednav. I will (yes). Ni vednav. I will not (no).

A wrîg ev môs dhô Benzans? Did he go to Penzance?

Gwrîg. He did. Ni wrîg. He did not.

’Esta ajŷ? Art thou at home?

Thoma. I am. Nynsov, or nynsoma, or (less correctly), nag ov. I am not.

In the case of a negative interrogative sentence the verb is immediately preceded by na=nî + a, whether it begins the sentence or is itself preceded by an interrogative conjunction. Thus:—

Na wrîsta gwelas? Didst thou not see?

Fraga na wrîsta crejy? Why didst thou not believe?

§ 3. Dependent Sentences or Subordinate Clauses.

These are of three kinds:—

1. Those introduced by conjunctions, such as if, that, as, etc., or by a relative pronoun.

2. Those analogous to the “accusative with the infinitive” of Latin.

3. The absolute clause.

1. The ordinary dependent clause introduced by a conjunction has its verb in the indicative, unless the so-called subjunctive is required to express uncertainty or contingency, without reference to any preceding conjunction. The verb is always in the simple inflected or inflected auxiliary form. The verb which follows the conjunction mar or mara, if, has its initial in the fourth state, and tre, tro, or dro, that, governs the second state.

A dependent sentence may sometimes precede its principal sentence, as in English. A very good instance of two sorts of dependent clauses may be seen in the following sentence from Boson’s Nebbaz Gerriau. The English is:—

“If that learned wise man [John Keigwin] should see this [i.e. this essay], he would find reason to correct it in orthography, etc.”

Boson’s Cornish, the spelling and division of words assimilated to that of the present grammar, is:—

Mar qwressa an dên deskes fîr-na gwelas hemma, If should [do] that man learned wise see this, ev a venja cavos fraga e ewna en scrîfa-composter. he would find why it to amend in writing-correctness.

In this sentence qwressa is for gwressa (third person singular of the conditional or pluperfect of the auxiliary gwîl, to do), with the initial in its fourth state after mar. Boson writes it markressa, all in one word. Fraga e ewna is an example of a variant of the second form of dependent sentence. The principal verb ev a venja cavos is in the impersonal auxiliary form, and of the two dependent clause verbs, one, qwressa an dên deskes fîr-na gwelas, is in the inflected auxiliary form, and the other, ewna, is infinitive.

In a relative sentence, if the relative pronoun is the subject, the verb appears to be in the impersonal form. That is to say, it is always in the form of the third person singular, and does not show any agreement with its antecedent, whatever person or number that may be in. The other peculiarities of relative sentences are given in Chapter VII. §4.

2. “Instead of using the conjunction that with another verb in the indicative mood, as in most European languages, it is usual to put the second verb in the infinitive preceded by the personal pronoun, as is common in Latin.” Thus says Norris, speaking in a manner perhaps rather less clear than usual, of an idiom found in the Dramas. This idiom, analogous to the “accusative with the infinitive” of Latin, is found down to the latest period of Cornish literature, though not to the complete exclusion of a finite clause beginning with that. The instances given by Norris are:—

Ha cous ef dhe dhasserhy, and say that he is risen.

Marth a’m bues ty dhe leverel folneth, I have wonder that thou shouldst speak folly.

Nyns a y’m colon why dhe gewsel, it goes not into my heart (i.e. I do not believe) that you have spoken.

Del won dhe bos, as I know thee to be.

Here are some later instances:—

Ny a wel an tîs younk dho e clappya leh ha leh, [164] we see that the young people speak it less and less (Nebbaz Gerriau).

Dre wrama crejy hedna dho bos gwîr yu serîfes enna, that I do believe that that is true that is written therein (Nebbaz Gerriau).

Nevertheless, one finds in the same piece:—

Ev a lavarras drova gever ǒl, he said that it was Goats All.

Bes mî a or hemma, dhort e hoer an Kernuak, drova talves bes nebbas, but I know this, by her sister the Cornish, that it is worth but little.

And in Keigwin’s translation of Genesis i.:—

Ha Dew a wellas trova da, and God saw that it was good.

A somewhat similar construction is sometimes used after dreven, because, and treba, until:—

Dreven tî dhô wîl hemma, because thou hast done this (Kerew’s Genesis, iii. 14).

Dreven tî dhô wolsowas dhô dalla dha wrêg, [165] because thou didst listen to the voice of thy wife (Gen. iii. 17).

Treba tî dhô draylya dhô’n nôr, until thou turn again to the earth (Gen. iii. 19).

Yet even there one finds

Dreven o hy dama a ŏl bewa, because she was the mother of all living (Gen. iii. 20).

Lhuyd mentions a similar construction after rag own, for fear, lest:—

Rag own whŷ dho gôdha po an rew dho derry ha whŷ dho vos bidhes, lest you fall or the ice break and you be drowned (literally, for fear you to fall or the ice to break and you to be drowned).

With fraga, why, one finds a similar form:—

Ev a venja cavos fraga e ewna, he would find why to amend it.

But when fraga introduces an interrogative sentence, an ordinary finite verb is used:—

Fraga (or rag fraga, “for why,”) na grejeth dhô’ m lavarow? Why dost thou not believe my words?

When “that” signifies “in order that,” the ordinary finite verb is used after it.

There is a peculiar construction, found chiefly in Jordan’s Creation, but also in the Ordinalia (e.g. Pass. Chr. 1120), for expressing “that I am.” It consists of the infinitive bos, to be, preceded by a possessive pronoun and followed by a pronominal suffix:—

Me a vyn mav fo gwellys ow bosaf Dew heb parow, I will that it may be seen, that I am God without equals.

And a still more confused one of the second person with the verbal particle y before bos, the pronominal suffix ta and the pronoun ge=dî:—

Me ny allaf convethas, y bosta ge ow hendas, I cannot understand that thou art my ancestor.

The first is analogous to the Welsh “infinitive construction,” as Rowland calls it, e.g. gwyr fy mod i yn dyfod, he knows that I am coming (lit. he knows my being in coming), only the Cornish form uses the pronominal suffix instead of the redundant personal pronoun.

3. The Absolute Clause. This construction, which answers more or less to the ablative absolute of Latin, and the genitive absolute of Greek, is common to all the Celtic languages. It is translated into English by a sentence introduced by when, while, whilst, or though, with a verb generally in the continuous form of the present or past tense, or by a participle. In the Celtic languages the absolute clause has two forms.

a. The affirmative, generally consisting of the conjunction and, a subject, noun or pronoun, and generally a participle. Rowland calls the conjunction, a, ac, of the Welsh form “the absolute particle,” and Professor Anwyl identifies it with a, ag, with, in an archaic form. But in Cornish ha or hag is used, and in Gaelic agus, and, in exactly the same way. The following are examples in Cornish, Welsh, and Gaelic:—

Cornish. An jy a ve gwarnes gan Dew, ha ’n jy ow cusca, [166] they were warned by God, and they sleeping, or, while they slept (Kerew’s translation of St. Matth. ii. 12, Gwav. MS.).

El a’n leverys dethy haneth, ha hy yn gwely pur thyfun, an angel said it to her this night, and she in her bed quite awake (Pass. Chr. 2202-4).

Welsh. Pa ham, a mi yn disgwyl iddi dwyn grawn-win, y dug hi rawn gwylltlon? Wherefore, and I looking to it to bring forth grapes [Auth. Vers., when I looked that it should bring forth grapes], brought it forth wild grapes? (Isaiah v. 4).

Gaelic. Do chonnaic Seaghán an duine, agus é ag teacht a-bhaile, John saw the man, and he coming home, i.e. when he was coming home.

b. The negative, in which not is expressed in Welsh and Cornish by heb, and in Gaelic by gan, both meaning without, followed by an infinitive:—

An delna ema stel ow tegy warnodha, heb wara dhodha teller vîth, [167] so it is still closing in upon it without leaving it any place (Boson’s Nebbaz Gerriau).

In many such cases this negative clause can be translated literally into English, and it is the usual form of negation with an infinitive or present participle.

A somewhat similar absolute clause of a descriptive character occurs occasionally:—

An golom, glas hy lagas, yn mes gura hy delyfre, the dove, blue her eyes, do set her free (Origo Mundi, 1105-6).

Un flogh yonk, gwyn y dhyllas, a young child, white his raiment (Passion, 254, 3).

In a similar construction in Welsh the adjective here agrees with the first noun, and the translation would be rather “The dove blue [as to] her eyes,” but in Cornish this is not so, for in this sentence golom (second state of colom) is feminine, so that the adjective would be las, not glas, if it agreed with it.

§ 4. The Infinitive or Verbal Noun.

The infinitive of a verb is treated almost exactly like a noun. If its object is a pronoun, this precedes the infinitive in the possessive form and governs its initial as it would that of a noun. If the object is not a pronoun, it follows the infinitive without change of initial, after the manner of an appositional genitive.

Very often the infinitive is governed by dhô, to, as in English, and under much the same circumstances, except that it is not so governed when it comes as the subject of another verb, and of course dhô is not used after auxiliary verbs. It is especially used after verbs implying motion.

Mî a vedn môs dhô ’gas gwelas, I will go to see you.

Mî eth dhô vetya an trên, I went to meet the train.

Lowen on ny dhô ’gas gwelas why, we are glad to see you.

When the sense of “to” is “in order to,” or the preceding verb implies an intention, the infinitive is generally preceded by rag or rag dhô, “for to,” or by a dhô, “of to.”

§ 5. Some Idioms and Expressions.

1. To have is expressed in three ways.

a. By the verb bos, to be, with the thing possessed as subject and the possessor in the dative form, i.e. preceded by dhô, to; cf. est mihi in Latin.

Affirmative. Ema levar dhem, there is a book to me.

Negative. Nynsyu levar dhem, there is not a book to me.

Interrog. ’Es levar dhem? Is there a book to me?

This is the common form in late Cornish.

b. By the verb cafos or cavos, to find, to obtain, used as an ordinary transitive verb with the possessor as subject and the thing possessed as object. This is not used for the present tense. Lhuyd gives a past tense, mî a gavaz or mî ’rig gavaz, I had, and a future, mî ven gavaz, I will have, but he, Norris, and Williams are all inclined to confuse this with the third form.

c. By a peculiar idiom compounded of a form of the verb bos, to be, and the third form of the personal (or else the possessive) pronouns. The explanation, as far as it goes, of this verb is to be found in Breton. Even there it has been confused a good deal, though its use is plain enough. Legonidec calls it “le verbe kaout [=Cornish cavos], avoir,” which he distinguishes from kavout or kaout, trouver; Maunoir, whose Breton, according to a picture in Quimper Cathedral, was received miraculously from an angel, wisely does not commit himself, but calls the verb, Latin fashion, after the first person singular of the present. Prof. Loth rightly speaks of it as “le verbe dit avoir,” and M. Ernault calls it “Verbe beza [to be] au sens de ‘avoir,’” and he explains it to be the verb to be, combined with the “pronoms régimes,” which is just what it is. In Breton it is not only used as the ordinary verb to have=to possess, but also as an auxiliary verb in the same manner as avoir, have, haben, are used in French, English, and German. This verb came to be used in Breton with or without the nominative pronoun being expressed. In Cornish the expressed nominative pronoun is less usual, except in the second person singular, where it is the rule. That it should be used at all in either language is a sign that in practice the original formation of the verb has been forgotten. Occasionally in Cornish this oblivion has resulted even in the application of pronominal inflections to the verb.

This form is found frequently in the Ordinalia and in the Poem of the Passion; it is fairly common in the Life of St. Meriasek, it is rarer in the Creation, and is not found at all in Cornish of the latest period (except in a doubtful and muddled form in Keigwin’s version of the Commandments), though Lhuyd gives a fragment of it in his Grammar, evidently taken from the earlier Dramas and not from oral tradition, for he takes the g of geffi and gefyth to be a hard g, whereas it is plainly a soft g for a d, as the analogy of tevyth, and of the Breton deveuz, devez, etc., shows. Moreover, it is sometimes written ieves, which is intended to represent jeves.

It will be well, by way of making this form clearer, to give not only the Cornish but also the corresponding Breton.

The tenses that are found are as follows:—

I. The Present.

Singular.

Cornish. Breton.

1. [mî] am bes [bus, bues, bues]. [me] em euz.

2. [tî] ath ĕs (thues). [te] ec’h euz.

3. m. [ev] an jeves (for deves). [hen] en deuz or deveuz.

3. f. [hy] as teves. [he] e deuz.

Plural.

1. [ny] an bes. [nî] hon euz.

2. [why] as bes. [c’houi] hoch euz.

3. [y] as teves. [hî] ho deuz or deveuz.

This tense is formed on us, eus, es (Breton euz), one of the forms of the third person singular of the verb substantive. To this is prefixed the verbal particle a, with the letter which is the third form of the personal pronoun, ’m, ’th, ’n, ’s, ’n, ’s, ’s, with the peculiar addition of jev and tev to the third persons and b to the others. The ’th of the second person singular is found written in this but not always in the other tenses, for it was probably often silent before f by a sort of assimilation. Its effect is observable in the initial mutation. Of this tense the first, second, and third persons singular and the second person plural are found. But for the existence of the form as bes [bues] for the last, one might suppose, with Williams, that the b of am bes was only the addition of a cognate letter to the m. But cf. the addition of b to oa and oe of the same verb in Breton.

II. The Future.

Singular.

Cornish. Breton.

1. [mî] am bedh (byth, beth). [me] em (or am) bez.

2. tî a [th] fedh (fyth). [te] ez (or az) pez.

3. m. [ev] an jevedh (for devedh). [hen] en devez.

3. f. [hy] as tevedh. [he] e devez.

Plural.

1. [ny] an (or agan) bedh. [ni] hor bez.

2. [why] as (or agas) bedh. [c’houi] ho pez.

3. [y] as tevedh. [hî] o devez.

It will be seen here and in the other tenses that the pronouns in Breton do not produce exactly the same mutations as in Cornish. The dh of Cornish is always written z in Breton, though that is pronounced dh in some dialects. The whole of this tense is found in the MSS.

III. The Preterite.

Singular.

Cornish. Breton.

1. [mî] am bê [bue]. [me] em (or am) boe.

2. tî ath fê. [te] ez (or az) poe.

3. m. [ev] an jeve. [hen] en devoe.

3. f. [hy] as teve. [he] e devoe.

Plural.

1. [ny] an (or agan) bê. [ni] hor boe.

2. [why] as (or agas) bê. [c’houi] ho poe.

3. [y] as teve. [hî] o aevoe.

Only part of this tense is found in the MSS., but the rest is easily formed by analogy.

IV. The Subjunctive (or Optative).

Singular.

Cornish. Breton.

1. [mî] am bo. r’ am bezo, bo.

2. tî ath fo, fetho. r’ az pezo, po.

3. m. [ev] an jevo (for devo, written gefo or geffo). r’ en devezo, devo.

3. f. [hy] as tevo. r’ e devezo, devo.

Plural.

1. [ny] an (or agan) bo. r’ hor bezo, bo.

2. [why] as (or agas) bo. r’ ho pezo, po.

3. [y] as tevo (written teffo, tefo). r’ o devezo, devo.

In this tense the Breton does not use the nominative personal pronoun, except when it is a form of the future, but prefixes r’ (ra). In Cornish re is used to make the optative and perfect, and in this case the ’th of the second person singular is not omitted, for re’ th fo and re ’th fê are the forms found.

A rather doubtful second tense (secondary present or imperfect), equivalent to the Breton am boa, may be conjectured in am beua (St. Mer. 47, 1686), am bethe may be the equivalent of the Breton imperfect subjunctive, am bize, bije, befe, and the third person singular of this may be the an geffa of St. Mer. 20, 159. Dr. Whitley Stokes gives both these forms as secondary presents. There is also a possible pluperfect te ny vea, and nyn gyfye, found in the second and third persons singular.

One finds such forms as am buef, as bethough, may ’stefons, etc., as instances of pronominal inflections added to this verb, showing how completely its derivation was forgotten, and it is further confused by being perhaps mixed up with the verb pewa (Welsh piau, Breton piaoua), to possess, a verb which in all three languages requires rather more disentangling than it has as yet received.

There are very full examples of this verb in Zeuss’s Grammatica Celtica (ed. 1871, p. 565).

2. Besides to have, certain other verbs are expressed with bos and the preposition dhô. Thus:—

Ma cov dhem [pron. ma códhem], I remember, lit. there is remembrance to me.

Ma whans dhem, I want, lit. there is want to me.

Ma whêr dhem, I am sorry, lit. there is grief to me.

Ma own dhem, I fear, lit. there is fear to me.

Ma dout dhem, I doubt, lit. there is doubt to me.

Ma reys dhem, or reys yw dhem, I must, lit. there is need to me.

Another expression for “to remember” is perthy cov, to bear memory. The imperative was sometimes written perco in one word. Perthy is used similarly with other nouns: na berth medh, be not ashamed, na berth own, be not afraid, na berth whêr, be not sorry, an vuscogyon orto a borthas avy, the fools hated him (Passion, 26, 3), na berth dout, do not doubt. The literal meaning is to bear shame, fear, sorrow, envy, doubt, etc.

Similarly nouns and adjectives are used with gan, with, as in Welsh, to represent states of mind. Thus:—

Da yu genev, I like, lit. it is good with me.

Drôg yu genev, I am sorry, lit. it is bad with me.

Gwell yu genev, I prefer, lit. it is better with me.

Marth yu genev, I am astonished, lit. wonder is with me.

Cas yu genev, I hate, lit. hate is with me.

The verbs dal and goth, signifying ought, it behoves, are used either impersonally or, though this is a late corruption, as ordinary verbs.

Ni dal dhen ny / Ni goth dhen ny } we ought not.

Or:—

Mî a dal / Mî a goth } I ought.

3. Gwyn an bês. This poetical expression is common to Cornish, Welsh, and Breton. It signifies, “fair the world,” i.e. happy, and is used with possessive pronouns and appositional genitives.

Gwyn ow bês, fair my world, happy I.

Gwyn dha vês, happy thou.

Gwyn e vês, happy he.

Gwyn bês an den na wrîg cerdhes en cŏsŏl an gamhin-segyon, blessed is the man who hath not walked in the counsel of the ungodly.

In Welsh, when the possessor of this “fair world” is expressed by a noun, there is a redundant possessive pronoun before byd (bês). Thus Psalm i. begins Gwyn ei fyd y gwr, fair his world of the man. But this is not the Cornish form, which uses the simple appositional genitive in such cases. There is a contrary expression, drôg pês, found in the Ordinalia (Passio Christi, 3089), drok pys of, unhappy am I. In this case drôg seems to put the initial of bês in its fourth state.

4. The following phrases are in common use, and are generally run into one or two words in pronunciation.

Mêr ’ras dhô Dhew (pron. merásthadew). Great thanks be to God.

Mêr ’ras dheugh why (pron. merásdhawhy). Great thanks to you.

Dew re dala dheugh why (pron. Durdladhawhy). God repay to you.

Dew re sona dheugh why (pron. Dursónadhawhy). God sain you.

Bennath Dew genough why (pron. Bénatew génawhy). The blessing of God be with you.

Dew genough why (pron. Dew génawhy). God be with you.

Pandráma (i.e. pa’n dra wrama). What shall I do?

Pandréllen (i.e. pa’n dra wrellen). What should I do?

Pándres (i.e. pa’n dra es). What is there?

Pandryu (i.e. pa’n dra yu). What is?

Pandresses (i.e. pa’n dra wresses). What shouldst thou do?

Fatla genough why (pron. fatla génawhy). How are you?

Trova (i.e. tre o-va), that he was.

§ 6. Rules for Initial Mutations.

1. The Second State.

a. A feminine singular or masculine plural noun (or adjective used as a noun) preceded by the definite article an, the, or the numeral idn, one, has its initial in the second state.

b. An adjective which follows and qualifies a feminine singular noun, has its initial in the second state.

c. A noun preceded by an adjective qualifying it, of whatever gender or number, has its initial in the second state.

d. If the adjective preceding and qualifying a feminine singular noun follows the article an, the, the initial of the adjective is also in the second state.

e. A noun in the vocative preceded by the particle a, O (expressed or omitted for the sake of verse), has its initial in the second state.

f. The possessive pronouns dha, thy, and e, his, are followed by words, whether nouns, adjectives, or verbal nouns (infinitives) in the second state. The form ’th, thee or thy, generally puts the word which follows in the second state, but sometimes in the fourth, or changes b to f, not v.

g. The verbal prefix ă (older y, yth), is generally followed by a verb in the second state.

h. The verbal particles a and re and the interrogative particle a are followed by a verb in the second state.

i. The prepositions a, der or dre, dhô, heb, re, and war, and compound prepositions ending in any of them, are followed by words in the second state.

k. The conjunctions tre, tro, that, pan, when, erna, until, hedre, whilst, are followed by the second state.

l. The adverbial particle en is followed generally by an adjective in the second state.

m. The adverbs pŭr, very, ni, na, not, fraga, why, fatla, how, are followed by initials in the second state.

2. The Third State.

a. The possessive pronouns ow, my, î, her, and aga, their, are followed by words in the third state.

b. Ma, may, that, are sometimes followed by verbs in the third state, and sometimes by a variant, g becoming h, and gw becoming wh.

3. The Fourth State.

a. The particle ow, which forms the present participle, is followed by a verbal noun (or infinitive) in the fourth state.

b. The conjunctions a, mar, mara, if, are followed by verbs in the fourth state.

c. The adverb maga, as (in “as well,” etc.) is followed by an adjective in the fourth state.

d. Sometimes an adjective beginning with d, when preceded by the adverbial particle en, has its initial in the fourth state, and rarely a noun beginning with d, when it follows in the appositional genitive a word ending in th.

e. The verbal prefix ă (y), when followed by verbs whose radical initial is d, often changes that initial to the fourth state, and in the case of those beginning with gw to wh. The conjunction ken, though, does the same.

f. The third form of the second personal pronoun singular ’th not infrequently changes the initial of a verb beginning with d to the fourth state, and that of one beginning with g or gw to wh. It also sometimes changes b to f.

The exact usage of the mutations is not very clear, for even the older writers used them rather wildly, but the above rules are the general principles of them. There are valuable notes on their phonetic principles in Dr. Whitley Stokes’s notes to St. Meriasek, and in a paper of additional notes which he published later. In the latest Cornish there was a tendency to use the second state after nearly anything, especially prepositions, except the few words which govern the other two mutations.