A Welsh Grammar, Historical and Comparative/Introduction

From Wikisource
Jump to navigation Jump to search

Introduction

Origin and general history

§ 1. i. The Welsh language is a member of the Keltic branch of the Aryan (also called the Indo-European or Indo-Germanic) family of languages.

The languages of this great family are classified as follows, names of branches and groups being printed in spaced type:

(1) Indian, comprising (a) Sanskrit; (b) Prākrit dialects, from which are descended numerous modern languages in India.

(2) Iranian: (a) Avestic (East Iranian, also called Zend or Old Bactrian); (b) Old Persian (West Iranian), later Pehlevi; (c) Modern Persian.

(3) Armenian.

(4) Greek, which comprises many dialects, the most important being (a) Ionic-Attic; (b) Doric; (c) Aeolic: Lesbian, Thessalian, Boeotian; (d) Arcadian and Cyprian; (e) Pamphylian.

(5) Albanian.

(6) Italic: (a) Latin, from which are derived the modern Romance languages; (b) Oscan, Umbrian.

(7) Keltic: (a) the Q division, consisting of dialects in Gaul and Spain, and the Goidelic group, comprising Irish, Scotch Gaelic and Manx; (b) the P division, consisting of Gaulish, and the British group, comprising Welsh, Cornish and Breton.

(8) Germanic: (a) Gothic; (b) the Norse group, including 1. Swedish, Gutnish, Danish; 2. Norwegian, Icelandic; (c) the West-Germanic group, including 1. Old English (or Anglo-Saxon), now English; Frisian; Old Saxon, now Low German; Dutch, Flemish; 2. Old High German, now German.

(9) Baltic-Slavonic: (a) the Baltic group: Old Prussian, Lithuanian, Lettish; (b) the Slavonic group: Old Bulgarian; Russian, Bulgarian, Illyrian; Czech, Sorabian, Polish, Polabian.

(10) Tocharish, recently discovered in East Turkestan.

ii. All these languages are descended from a common ancestor called the Aryan parent language, Primitive Aryan, or briefly Aryan. Similarly, the languages of each branch may be referred to a common parent called Primitive Keltic, Primitive Italic, Primitive Germanic, etc., as the case may be. Some of the above branches are perhaps to be regarded rather as groups; Indian and Iranian are often classified together as the Indo-Iranian branch; and the common features of Keltic and Italic are such as to render it certain that the two branches were united and shared the same development for a period after their separation from the others; hence we may classify them together as Italo-Keltic; see § 86 ii (2), § 113 i (3), § 147 iv (2), § 203 vii (3).

iii. Our earliest knowledge of the various languages varies widely in point of date, and naturally those of which we possess the most ancient records on the whole bring us nearest the fountain head. But the Baltic group, of which our knowledge is only recent, are of a remarkably archaic character; Lithuanian, whose earliest text is dated 1547, and which has changed comparatively little since, preserves to this day some forms which are practically identical with those which we have to postulate for Primitive Aryan itself.

From the cradle of Aryan speech various tribes migrated at different periods in different directions, establishing themselves in distant lands, in which their speech prevailed, though the aborigines cannot have been exterminated, since the speakers of Aryan languages in historical times belong to many races, and it is still matter of dispute which of these has the best claim to be regarded as representing the original Aryans. The dispersion commenced not earlier than about 2000 b.c. according to Hirt, Die Indogermanen 22. The centre of dispersion is now generally believed to have been somewhere in Europe.

A parent language is not necessarily isolated; analogy rather suggests the contrary. As Latin, which is the parent of the Romance languages, is derived from Aryan and allied to the other Aryan languages, so Aryan itself must be derived from some remote ancestor, and it is improbable that it is the only descendant of it which survived. Sweet, by a comparison of the pronominal and verbal forms of Aryan and Ugrian, has made out a strong case for supposing that the two families are allied; see his History of Language pp. 112 ff. On the other side Moller, in his Semitisch und Indogermanisch i (1907), has compared the consonant sounds of Aryan in detail with those of Semitic, and in KZ. xlii 174 ff. the vowels; and claims to have proved their derivation from a common source. But none of these affinities can yet be regarded as established.

§ 2. In the oldest forms of Goidelic found in the ogam inscriptions, Primitive Keltic q from Aryan q remains; but in the oldest British it had already become p, and it is p in Gaulish. Traces of a Keltic q language in Gaul are seen in names like Sequani; and in some recently discovered inscriptions further evidence of the survival of such a language is believed to have been found. As the change of q to p is the earliest soundchange known which is not common to the whole branch, it seems reasonable to classify the Keltic languages as above § 1 i (7).

The more usual classification adopted in recent years is that in which the Keltic languages are grouped into "insular" and "continental". But this is a negation of all classification; it is as if we were to group together English and Icelandic as insular Germanic! Thurneysen now calls it a "geographic" classification (Gr. 1), which is equivalent to saying that it is no classification at all. It arose out of the view put forward in Rhys's LWPh.² (1879) pp. 16 ff. that the language of the ogam inscriptions in Wales is an old form of Welsh. Thurneysen, KR. (1884) pp. 7 ff., adopts this view; dismisses Rhys's later view, CB. (1884) p. 215, that the ogams are Irish; and concludes that, as the ogams have q, the change q > p in British is much later than the same change in Gaulish. Of course, if the ogams are Welsh, there was no difference in the 5th cent. between Welsh and Irish, and both differed from Gaulish, which alone had p. Hence the classification into insular and continental. But the assumption on which it is based is groundless; no one now holds that the ogams are Welsh.

If it is denied that a systematic classification of the groups is possible, it would be better to take them separately than to adopt a classification which implies a close relationship between Goidelic and British. But there seems no sufficient reason for separating British from Gaulish. It is now admitted that Brit. p from q is ancient; and it is extremely improbable that this p developed independently of Gaulish p. Tacitus, Agricola xi, tells us that the speech of the Britons differed little from that of the Gauls. The Gaulish forms Πεννο-ουινδος, Vindomag(os), ambact(os), Voretovir(os) are identical with the British forms which we have to postulate as the originals of the Welsh penwyn 'white-headed', gwynfa 'paradise', amaeth 'serf', gwaredwr 'saviour'. It is for those who would separate British and Gaulish to prove that Tacitus was wrong.

For the continental q dialect or group of dialects various names have been suggested, as Sequanian (Nicholson), Pictavian, Celtican (Rhys), Ligurian (Jullian). The language of the Coligny calendar contains both qu and p; but whether the latter is secondary, or borrowed from Gaulish, or represents Aryan p, cannot yet be decided, since independent evidence as to meaning is lacking. The presence of Ar. p, if proved, would constitute these dialects a class apart.

§ 3. i. Welsh, Cornish and Breton are descended from British (properly Brittish), the language of the ancient Britons. The speakers called themselves Brittones, and their language *Brittonikā.

The Old English name was Brittisc or Bryttisc, as On Bryttisc sprecende Guthlac, Godw. 42, 17 (cf. Rhys, CF. 676), which in later spelling was Brittish, misspelt British[1] under the influence of the Lat. Britannia. The name continued to be used for the derived languages: “The Gaulish speach is the very Brittish, the which was very generally used heere in all Brittayne before the coming in of the Saxons; and yet is retayned of the Walshmen, the Cornishmen, and the Brittons,” Spenser, State of Ireland (Lloyd’s Enc. Dic.). It was commonly used for Welsh as late as the 18th and beginning of the 19th cent.: “In these Schools…​Men, Women and Children being ignorant of the English Tongue, are taught to read their native British language,” Welsh Piety 1754 p. 53, 1755 p. 47 etc. Cf. dedication of Grawn Awen (Caledfryn) 1826.

ii. The Welsh call themselves Cymry, from *kom-brogī ‘fellow countrymen’; but the use of this as a national name is subsequent to the separation of the Welsh from the Cornish and the Bretons. The old name, which survived in poetry, was Brython b.t. 13 from Brittones; the corresponding name of the language Brythoneg was superseded by Cymraeg, but some memory of it survived (D.D. gives Brythoneg, but with no quotation). The Bretons call their language Brezonek, and Cornish was called Brethonec; all these forms imply an original *Brĭttŏnĭkā. Sir John Rhys in his LWPh.² 16 adopted the names Brythons and Brythonic for the Brittones and their language, remarking, however, that he would “like to have called them Brittons and their language Brittonic”. I prefer to call the language by its traditional English name British, which in this connexion involves no ambiguity. The term Brythonic suggests a later period, and tends to disguise the fact that the language meant is the speech of the ancient Britons.

iii. The name Brittō, sg. of Brittones, probably owes its tt to its being a formation of the type of Gk. Νικοττώ etc., see § 93 iii (2), for an earlier Britann(os), pl. Britannī. Similarly we have a late Βριττία for Britannia. *Brittia survives in Bret. Breiz ‘Brittany’, and *Britanni̯ā, in Ml. W. Brydein used as a variant of the more usual Prydein as in b.b. 100, milvir Pridein l. 5, milguir Bridein l. 7. Britan- seems to be for Pritan- by British alternation p:b § 101 iii (2); cf. prit(an)nii Holder i 564, pritnnii do. ii 1046. Pritto also occurs as a personal name beside Britto, and Prittius beside Brittius (see Holder s. vv.). The view now generally held that the members of these pairs are unrelated rests on no other basis than the assumption that British p- could under no circumstances pass into b‑. The fact, however, is that Pritan- and Britan- are synonymous. The P- goes back through Diodorus Siculus probably to Pytheas (4th cent. b. c.). Polybius (2nd cent. b. c.) seems to have used Βρετ(τ)ανικαὶ νῆσοι; but Strabo and Diodorus have Πρετ(τ)ανικαὶ νῆσοι and Πρετ(τ)ανοί; later Ptolemy and Marcian used Π‑. Stephanus of Byzantium (c. a. d. 500) wrote Βρετ(τ)ανίδες νῆσοι and Βρετ(τ)ανοί, remarking that Dionysius (Periegetes; Augustan age) wrote “one t…​Βρετᾱνοί” [read Βρεταννοί], and that others used “p, Πρετανίδες νῆσοι, as Marcian and Ptolemy”; elsewhere Stephanus himself wrote Πρετανίκή and Πρετανοί Holder i 560. The ε in Πρετ- ≡ Brit. ĭ, see § 66 i. Pritan- is an n-stem representing original (*qr̥tₑn- or) *qritₑn‑; for the nn see § 62 i (2).

The surviving forms show that the old P- forms had one t; thus W. Prydain ‘Britain’, Ml. W. Prydein, implies *Pritan(n)i̯a and Ir. Cruithnech ‘Pictish’ implies a Pictish *Pritenikos; hence the ‑ττ- in Πρεττανικαὶ is probably a misspelling of copyists, due to the Britt- forms which prevailed later. The forms with ‑on- had ‑tt‑; thus W. Brython < Brittones, Bret. Brezonek < *Brittonikā, and Ml. Ir. Bretain 'Britons' represents Brittones regularly. As the new form Brittones spread, Britannia became Brittannia which survives in Fr. Bretagne; later we find Brittania Βρεττανική etc. which were substituted for older forms in mss. There is no possible doubt that the oldest B- form is Britann‑: Catullus (died 54 b. c.), Propertius, Vergil, Horace, Ovid, all scan Brĭtann‑. The evidence of the dated coins and inscriptions in Holder is as follows (the numbers in brackets refer to Holder i): coins of Claudius a. d. 41, 46 have Britannis, Britanni., Britann. (564, 36, 37); inscriptions: a. d. 41 Britannia (589, 52); time of Claudius Britannia (590, 27); a. d. 43 Britannic(um) (598, 24); a. d. 49 Britan(nicis) (599, 34). In a. d. 49 or 50, at least a century after the first evidence of Britann‑, ‑tt- appears first in two inscriptions in the name of Claudius’s son Britannicus: Britta[nico] (602, 18), Brittanici (602, 22); in eight other cases it is Britannicus or Βρεταννικος (602). The early appearance of tt in this name may mean that Brittō was in use as an abbreviated personal name earlier than as meaning ‘Briton’. In the national name the single t continued in use: a. d. 54 Britan(nicum) (600, 22); a. d. 65 Britannico (599, 5); a. d. 80 Britannica (598, 37). In a. d. 85 Brittones first appears in the gen. pl. Brittonum side by side with Britannica (607, 41–2). In a. d. 90 first occurs Brittanniae (588, 7); in a. d. 98 and 103 Britannia again (590, 25; 588, 9); in a. d. 99 Brittonum (607, 43); in a. d. 105 Brittan[nia] (588, 10), in a. d. 110 Brittanniae (590, 5) and Britannica (598, 40). In the 2nd cent. Britann- and Brittann- are both common. Brittania first occurs on a coin in a. d. 185 (590, 50) and Brittanicae in a. d. 210 (599, 51).

The W. Pryden ‘Picts’ § 121 iii from *Pritenes, Prydyn b.t. 13 ‘Pictland’ from *Pritenī, and the Ir. Cruithen Cruithnech seem to have the F-grade ‑en- of the stem-ending, probably a Pictish form. The Picts were Britons, as shown by the fact that p < q abounds in Pictish names. They kept in their own name the P- which also survives in W. Prydain 'Britain', and so came to be distinguished from the Southern Britons, who called themselves Brittones. Picti, which is not known to occur before a. d. 297, seems to be a Latin translation of *Pritenes explained as meaning 'figured' (: W. pryd 'form', Ir. cruth), just as W. Brithwyr 'Picts' is a translation of Picti. This explanation of *Pritenes is probably only a piece of popular etymology; but even if it had some old tradition behind it, the name is equally applicable to the other Britons, for they all painted or tattooed themselves, Caesar b.g. v 14, Herodian iii 14, 7. Indeed the objection to accepting it as the true explanation is that at the time when it was first applied it could not be distinctive.

The etymology of a proper name is always uncertain, except when, like Albion, it hardly admits of more than one meaning, and that meaning fits. Britain like Albion must have been a name given to the island by its Keltic invaders, and Albion suggests the feature most likely to impress them. There is an Italo-Keltic root of some such form as *qrēi- which means 'chalk' or 'white earth', giving Lat. crēta, and W. pridd 'loam', Irish crē; the attempt to derive the Welsh and Irish words from the Latin is a failure​—​the root must be Keltic as well as Italic; and it may have yielded the name Pritannia meaning 'the island of the white cliffs'.

§ 4. i. Gaulish and British are known to us through names on coins, and words and names quoted by Greek and Latin authors. No inscriptions occur in British, but British names are found in Latin inscriptions. A number of inscriptions in Gaulish have been preserved. Goidelic is known from the ogam inscriptions, of which the oldest date from the 5th century.

ii. The scanty materials which we possess for the study of Gaulish and British are sufficient to show that these languages preserved the Aryan case-endings, and were at least as highly inflected as, say, Latin. The great change which transformed British and converted it into Welsh and its sister dialects was the loss of the endings of stems and words, by which, for example, the four syllables of the British Maglo-cŭnos were reduced to the two of the Welsh Mael-gwn. By this reduction distinctions of case were lost, and stem-forming suffixes became a new class of inflexional endings; see § 113, § 119 i.

§ 5. The history of Welsh may be divided into periods as follows:

(1) Early Welsh, from the time when British had definitely become Welsh to the end of the 8th century. Of the forms of this period we have only echoes, such as the names found in Bede, § 113 i (4).

(2) Old Welsh (O. W.), from the beginning of the 9th to the end of the 11th century. The remains of this period are a number of glosses, and some fragments of prose and anonymous verse. But O. W. forms are preserved in later copies in the genealogies, the Book of Llandâf, the Laws, the Book of Aneirin, etc.

(3) Medieval Welsh (Ml. W.), from the beginning of the 12th to the end of the 14th century and somewhat later. The orthography varied much during this period, and was at first in an unsettled state. It will be convenient to refer to the language of the 12th and early 13th century as Early Ml. W., and to that of the 14th and early 15th as Late Ml. W.

(4) Modern Welsh (Mn. W.), from Dafydd ap Gwilym to the present day. Though D. ap Gwilym wrote before the end of the 14th century, he inaugurated a new period in the history of the language, and is in fact the first of the moderns. The bards of the 15th and 16th centuries wrote the bulk of their poetry in the cywydd metre popularized by Dafydd; and the forms used by him, with some alterations of spelling (ai, au for ei, eu § 79), were preserved unchanged, having been stereotyped by the cynghanedd. The language of this body of poetry may be called Early Mn. W.

At the introduction of printing, Wm. Salesbury attempted in his works, including the New Test. (1567), to form a new literary dialect, in which the orthography should indicate the etymology rather than the sound. His practice was to write Latin loanwords as if no change had taken place in them except the loss of the ending, thus eccles for eglwys 'church', descend for disgyn ‘to descend’; any native word with a superficial resemblance to a Latin synonym was similarly treated, thus i ‘his, her’ was written ei because the Latin is eius (perhaps eu ‘their’ suggested this). But Dr. Morgan in his Bible (1588) adopted the standard literary language as it continued to be written by the bards, though he retained some of Salesbury’s innovations (e.g. ei for i ‘his’). Some dialectal forms used by Morgan (e.g. gwele for gwelai ‘saw’ § 6 iii) were replaced by the literary forms in the revised Bible (1620), which became the standard of later writers. Thus Late Mn. W., which begins with the Bible, though influenced to some extent by Salesbury, is based upon Early Mn. W., and forms a continuation of it. In the 19th century several neologisms were introduced, chiefly under the influence of Pughe; the language of this period will be referred to, when necessary, as Recent Welsh.

§ 6. i. The spoken language has four main dialects, as follows:

(1) Venedotian, the dialect of Gwynedd or North West Wales. (Gwyn. dial.)

(2) Powysian, the dialect of Powys, or North East and Mid Wales. (Powys dial.)

(3) Demetian, the dialect of Dyfed or South West Wales.

(4) Gwentian, the dialect of Gwent and Morgannwg, or South East Wales.

N. W. is used as an abbreviation for 'North Wales' or 'North Walian', S. W. for 'South Wales' or 'South Walian'.

ii. The two N. W. dialects differ from the two S. W. chiefly in the choice of words to express some common ideas, the most noticeable difference being the use of o, fo in N. W., and e, fe in S. W., for the pronoun 'he' or 'him'.

iii. In the final unaccented syllable the diphthongs ai and au are mostly levelled with e in the dialects. In Powys and Dyfed, that is, in an unbroken belt from North East to South West, the three are sounded e; thus cader, pethe, bore for cadair 'chair', pethau 'things', bore 'morning'. In Gwynedd and Gwent they are sounded a, as cadar (Gwent cá|tar), petha, bora. When ai is significant (e.g. as denoting the plural) it is ai in Gwynedd, i in Gwent, sometimes i in Powys, as Gwynedd defaid 'sheep', llygaid (when not ll'gada) 'eyes'; Gwent defid, llý|cid; Powys defed, llygid; Dyfed defed, llyged.

Dialectal forms, chiefly Demetian and Powysian ‑e, begin to appear in the mss. of the 15th century; but the rhymes of the bards of the 15th and 16th centuries, with the exception of some poetasters, always imply the literary form, which is still used in the written language except in a few words. See § 31 ii.

Notes
  1. It is of course still pronounced Brittish, rhyming with skittish, not with whitish.