STOP Early Journal Content on JSTOR, Free to Anyone in the World This article is one of nearly 500,000 scholarly works digitized and made freely available to everyone in the world by JSTOR. Known as the Early Journal Content, this set of works include research articles, news, letters, and other writings published in more than 200 of the oldest leading academic journals. The works date from the mid-seventeenth to the early twentieth centuries. We encourage people to read and share the Early Journal Content openly and to tell others that this resource exists. People may post this content online or redistribute in any way for non-commercial purposes. Read more about Early Journal Content at http://about.jstor.org/participate-jstor/individuals/early- journal-content . JSTOR is a digital library of academic journals, books, and primary source objects. JSTOR helps people discover, use, and build upon a wide range of content through a powerful research and teaching platform, and preserves this content for future generations. JSTOR is part of ITHAKA, a not-for-profit organization that also includes Ithaka S+R and Portico. For more information about JSTOR, please contact support@jstor.org. ON CORREPTION IN HIATUS By Edward Bull Clapp It is a familiar fact that in Greek hexameter poetry diphthongs and long vowels at the end of a word, followed in the same verse by a word beginning with a vowel, are usually treated as short (vocalis ante vocalem corripitur). But the explanation of this correption is uncertain, its relation to the tolerance of hiatus is obscure, and even the elementary facts concerning the extent and proportions of the usage itself, the vowels and diphthongs in which it appears most frequently, and the difference in practice among the different poets, are loosely stated in the most recent handbooks. It has therefore seemed worth while to make a new and somewhat extended examination of this feature of dactylic verse, covering a considerable number of poets from the earliest to the latest times. Such an examination may furnish us at least with a clear general view of the extent of the practice and the cir- cumstances under which it occurs, and at the same time may shed some light on the difficult problem of its origin and explanation. The poets whose usage has been noted are twenty-six in num- ber, and the list includes the chief representatives of the epic, classical, Alexandrian, Roman, and Byzantine periods. Certain poets of great intrinsic importance are passed over because we possess too little of their work to afford a sufficient basis for obser- vation. The parts of Homer examined were chosen from among the earliest, as well as the latest, portions of the poems, in the hope of finding such a difference of usage as should be significant. But no such marked difference was found to exist. Homeric cor- reption is fairly constant from A to a>, and the slight differences have no apparent significance for our inquiry. In most cases at least 1,000 consecutive verses of each poet were examined, 1 or the 1 It was found that the practice of each poet, in regard to correption, was reason- ably uniform, and that nothing of importance was likely to be gained by including more than 1,000 verses in the study. [Classical Philology I, July, 1906] 239 240 Edward Bull Clapp whole extant remains of poets from whom a less number of hex- ameters are preserved. The few exceptions to this rule will be seen in the following table, which shows in detail the portions of each poet on which the later tables are based: Verses Homer, A, B 484-877, K, X, t, w [Dindorf-Hentze] 3,213 Hesiod, Theogony 1,022 Homeric Hymns to Del. and Pyth. Apollo, and Hermes 1,126 Tyrtaeus 145 Solon 220 Theognis 1,430 Simonides 364 Empedocles 419 Theocritus, Idylls i-xxii 2,062 Aratus 1,154 Manetho, ii [i], iii [ii] 930 Callimachus 1,356 Apollonius Rhodius i 1,362 Moschus 229 Bion 209 [Phocylides] 230 Nicander, Theriaca 958 Oppian, De venatione i, ii 1,166 Quintus Smyrnaeus i, ii 1,496 Dionysius Periegetes 1,186 [Orphic] Argonautica 1,373 Nonnus, Dionysiaca i, ii 1,246 Colluthus, Helen 385 Musaeus, Hero and Leander 341 Tryphiodorus 673 Tzetzes, Antehomerica, Homerica 1-194 600 The following table shows the frequency of correption of the several diphthongs and long vowels in the different poets. The figures indicate the average number of instances to each 100 verses of text. The vowels and diphthongs are arranged in the order of prevailing frequency in the whole list of poets. This may aid the reader's eye to interpret the results, since the figures opposite each poet's name will in general diminish from left to right, so that any marked exception to this rule (as in the case of Theocritus) indicates a practice more or less unusual. The diph- thongs ev and av are occasionally shortened, but they are omitted here to save space: On Coeeeption in Hiatus 241 Homer. , Hesiod , Homeric Hymns Tyrtaeus Solon Theognis Simonides Empedocles Theocritus Aratus Manetbo Callimacims Apollonius Rhodius . . . Moschus Bion [ Phocylides ] Nicander Oppian , Quintus Smyrnaeus . . . Dionysius Periegetes . [Orphic] Argonautica. Nonnus Colluthus Musaeus Tryphiodorns Tzetzes at Ol ov €1 Ct> 1 V a a 16 7.3 1.5 1.6 1.5 0.8 0.6 0.7 0.1 19 4.1 4.7 1.7 1.9 1.2 1.1 0.1 0.0 18 6.0 3.0 1.2 1.5 1.3 0.7 0.1 0.0 10 4.0 2.0 1.3 1.3 0.0 0.0 1.3 0.0 12 2.3 1.4 1.9 0.5 1.0 0.5 0.0 0.0 16 4.5 1.4 1.5 0.6 1.6 1.0 0.6 0.1 4 1.4 2.2 0.3 2.7 0.3 0.5 0.5 0.5 24 3.6 0.7 2.4 0.5 0.8 0.2 0.2 0.0 17 6.0 0.7 1.9 1.9 0.5 0.3 2.1 1.5 17 6.4 1.9 1.5 1.9 0.9 0.5 0.1 0.0 IS 3.2 2.6 1.2 1.7 2.4 0.6 0.0 0.0 12 3.1 0.4 0.4 0.4 1.0 0.0 0.2 0.0 15 6.3 2.2 1.1 1.0 0.7 0.3 0.5 0.1 15 6.0 0.4 0.8 0.4 0.8 0.0 0.0 0.0 17 3.3 0.0 1.0 0.5 0.5 0.0 0.0 1.0 10 2.6 1.3 0.9 0.4 0.0 1.3 0.0 0.0 26 3.3 2.9 1.3 1.5 1.1 0.2 0.1 0.0 7 0.7 0.5 0.1 0.1 0.1 0.1 0.2 0.0 19 7.7 2.3 0.8 0.5 1.0 0.5 0.6 0.0 21 4.6 1.7 1.0 0.7 0.5 0.3 0.0 0.0 11 4.5 1.6 0.8 1.2 0.7 0.4 0.4 0.0 12 1.0 0.0 0.0 0.1 0.2 0.1 0.2 0.0 16 0.3 0.3 0.0 0.0 0.0 0.0 0.3 0.0 11 3.2 0.3 0.6 0.0 1.2 0.0 0.3 0.0 13 1.8 0.0 0.0 0.1 0.0 0.0 0.0 0.0 13 3.5 2.0 0.2 1.7 1.1 0.3 0.3 0.0 0.0 0.0 0.0 0.0 0.0 0.1 0.5 0.0 0.5 0.0 0.0 0.1 0.0 0.0 0.0 0.0 0.1 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 It will be seen from the above table that the diphthong ai is shortened, not only more frequently than any other single vowel or diphthong, but actually more freqently than all the other vow- els and diphthongs combined. Of this very large number of cases of the correption of at, about three-fifths occur in the single word Kai, while most of the remaining instances appear in the inflectional endings of verbs, especially -rat and -fwu, though there are a few nominatives plural. But it would appear that more than one-third of all the cases of correption in hiatus in Greek poetry are furnished by Kai alone. The following table shows the total number and distribution of the instances of shortened at, adding in detail the practice of certain poets who display a marked divergence from the prevailing tendency. In general, the preponderance of ai over the other vowels and diphthongs is most overwhelming in Oppian, Nonnus, Colluthus, and Tryphio- dorus, and in all of these but Nonnus the vast majority of cases appear in icaC. In this connection it is significant that Kai is practically never elided, although «e often suffers elision and he constantly. 242 Edward Bull Clapp KOI Verbal Ending Scattering 2,266 89 180 111 56 1,487 128 22 110 149 3 191 53 32 The diphthong ot is shortened only one-fourth as often as at. The nominative plural furnishes almost exactly one-half of all the cases of short ot, and most of the remainder occur in the datives ol, not,, poi. In view of the fact that the accentual law for final oi is the same as for final ai {avdpwirot, OaXaa-aai), the different behavior of the two diphthongs in hiatus is remarkable. The fol- lowing table shows (1) what may be called the natural frequency of final oi and ac as seen in Thucydides and Xenophon; (2) the frequency of correption in hiatus (in hexameter) in the same diphthongs; (3) the frequency of correption in the nominative plural. The figures represent the average number to 100 lines of text. at 01 63 15 0.7 34 4— 2— The shortening of ov occurs only in the genitive singular. The diphthong ei is shortened most frequently in the word iiret, and in the third person singular of verbs. Of the "improper" diph- thongs, a> and a are shortened only in the dative singular. The same is true of »? in the great majority of cases, though a few sub- junctives appear. Of the long vowels, 17 appears as short chiefly in nominatives singular, and in the particles Srj, pr), fj, and others, although the list in Theognis is much lengthened by that poet's frequent use of the vocative YloXwatSr) ( w ~ — ~). Correption of o) is seen most often in iyd>, and in the first person singular of verbs, except in Theocritus, where the many genitives in -mvov to o-vvwktov avras [i. e., final and initial vowels], icc^voras direpya^ojucvai toiis r}x ov ' i > T V V T 5 S uivrji, the termination -fu, and elsewhere. The conjunction /cat, to be sure, which furnishes so very large a pro- portion of the cases of metrical correption (see p. 242), is seldom, if ever, elided; but this fact need not be interpreted as unfavor- able to the "semi-elision" theory. On the contrary, it may be regarded as distinctly supporting the opinion of Curtius. On Cokkeption in Hiatus 245 W. Hartel, in the work referred to above (p. 243), noticed that correption in hiatus is most frequent in diphthongs consisting of a short vowel followed by i, and on this fact he based a new effort to explain the practice. Believing that the usage began with these diphthongs, he assumed that in such combinations as koI iyd, the final i of the diphthong, standing between two vowels, under the influence of the rapid movement of dactylic verse, passed over into the corresponding consonantal sound of jot. The emergence of this consonantal sound at once removed the hiatus, and the pre- ceding -a-, at the same time, being left by itself, appeared with its natural short qtiantity. Thus the verse (A 437) «« Si koX avrol fiaZvov tin pryyfuvi Oa\d = Kaj ejw) to the termination -ov of the genitive (e/c TlvXov ekOmv = iic Ilv\of eXBdiv) . Allen rightly calls attention to the fact that -ov in the genitive was never a real diphthong, so that it is im- possible to speak of the "consonantization" of the second element. The handbooks of metric give us little assistance. Christ (p. 38) says simply: 1 Grulich does not, however, mention the Boeotian -cu for -p, which may have had an influence on the shortening of the latter. 246 Edward Bull Clapp Von einem Hiatus kann nicht mehr die Eede sein wenn .... ein langer Vocal .... in der Thesis verkiirzt wird . . . . , denn so wurden eben regel- massig die Vocale an den bezeichneten Stellen von den griechischen Schriftstellern behandelt. Elsewhere (p. 26), however, he seems to endorse the views of Hartel and Grulich, and says : Der Grand, dass die Kurzung gerade in der Thesis des Dactylus eintrat, lag in der raschen rollenden Aussprache, welche die Aufeinan- derfolge mehrerer Kiirzen mit sich brachte, und durch welche nament- lich die Diphthonge sich leicht in ihr rein vocalisches und ihr halb- vocalisches Element aufldsten. Westphal (III. i, p. 122) combines the principle of "semi- elision" with the Hartel-Grulich consonantization of t and v. After comparing correption in hiatus with elision, he says: Wie dort [i. e., in elision] die einzeitige kurze Vocal zur zeitlosen Vor- schlagsilbe wird, so verliert hier [i. e., in hiatus] der zweizeitige lange Vocal die Halfte seines Werthes, und wird zur einzeitigen Kilrze. Be- sonders haufig werden von dieser Verkiirzung die diphthongischen Auslaute, at, «, ot, und ov [but see p. 241] betroffen, deren zweiter Be- standtheil, das t oder v, hierbei in einen Halbvokal tibergeht. It thus appears that, while the convenient term "semi-elision" has found considerable favor as describing correption in hiatus, yet the consonantization theory of Hartel and Grulich has also been widely accepted as affording a more fundamental explana- tion of the process. It is at this point that the hope of more light from a wider field of investigation presents itself. As sug- gested by Allen in the paper cited (p. 121), if it should be found that after Homer the correption shows a tendency to spread more and more from the "short" diphthongs to the "improper" diph- thongs and the simple long vowels, it would add considerably to our confidence that Hartel and Grulich were right ; that the prac- tice began with the diphthongs ai, ei, ot, ev, as explained by those scholars, and that its appearance in other vowels and diphthongs is to be accounted for either by analogy or in some other way. The results in detail have been set forth in the table on p. 241. But for the sake of clearness and sharpness of presentation, the On Correption in Hiatus 247 following table is added, in which those cases of correption which are satisfactorily explained by Hartel and Grrulich are combined into one group ("A"), while those cases which represent the ten- dency toward the extension of the field of correption form another group ("B"). The diphthongs eu and av are ignored as before. In 100 Verses Homer Hesiod Homeric Hymns Tyrtaeus Solon Theognis Simonides Empedocles Theocritus Aratns Manetbo Callimachus Ap. Rhodius Moschus Bion [Phocylides] Nicander Oppian §. Smyrnaeus ion. Periegetes [Orph.j Argonautica Nonmis Colluthus Musaeus Tryph iodorus Tzetzes Per cent. Total A B A 30.1 24.9 5.2 0.83 33.8 24.8 9.0 0.73 31.8 25.2 6.6 0.79 19.9 15.3 4.6 0.77 19.6 16.2 3.4 0.83 27.4 22.0 5.4 0.80 12.9 5.7 7.2 0.44 32.4 30.0 2.4 0.93 32.7 24.9 7.8 0.76 30.2 24.9 5.3 0.82 24.7 17.4 7.3 0.70 17.6 15.5 2.1 0.88 27.2 22.4 4.8 0.82 23.4 21.8 1.6 0.93 23.3 21.3 2.0 0.91 16.5 13.5 3.0 0.82 36.5 30.6 5.9 0.84 8.8 7.8 1.0 0.89 32.4 27.5 4.9 0.85 29.8 26.6 3.2 089 20.6 16.3 4.3 0.79 13.6 13.0 0.6 0.96 16.9 16.3 0.6 0.96 16.6 14.8 1.8 0.89 14.9 14.8 0.1 0.99 22.1 16.7 5.4 0.76 0.17 0.27 0.21 0.23 0.17 0.20 0.56 0.07 0.24 0.18 0.30 0.12 0.18 0.07 0.09 0.18 0.16 0.11 0.15 0.11 0.21 0.04 0.04 0.11 0.01 0.24 The above table affords little evidence of steady progress, from Homer onward, in the direction of widening the field in which correption takes place. On the contrary, the practice of Homer is seen to be very nearly the mean or normal usage of the whole list. Hesiod, to be sure, shows a decidedly larger proportion of the less usual cases of curtailment of quantity (B) than Homer, but after Hesiod we find no further movement toward " B," of any consequence, except in Simonides. In this poet alone the cases under "B" are actually more numerous than those under "A," though this is due, not to a numerical increase in the former class, but to a marked falling-off in the latter. On the other hand, in the Homeric Hymns, Tyrtaeus, Solon, and Theognis (the last three, like Simonides, employing the elegiac couplet), the proportion is much the same as in Homer, while in Empedocles there is a 248 Edwabd Bull Clapp decided reaction toward "A." The Alexandrian poets do not differ materially, in their usage, from their earlier models, except that in Moschus and Bion the tendency toward confining correption to the "A" diphthongs becomes still more marked. The poets of the Roman and Byzantine periods differ widely among themselves, but from Nonnus onward there is comparatively little correption of any kind, and the little that appears is confined more closely than ever to the diphthongs under "A." Tzetzes forms the only exception, and approaches very closely to the Homeric norm. It would be interesting to know whether this was studied, or uncon- scious, imitation of his great prototype. In short, what little progress can be observed, from Homer to Tryphiodorus, lies in the direction of narrowing the range of cor- reption, rather than of extending it. The opinion of Hartel and Grulich does not therefore receive the kind of confirmation which was hoped for (see p. 246). If correption began with the "short" diphthongs, and extended to the other diphthongs and long vowels by analogy, the process must have begun far back of our Homer, and reached its maximum effect in Hesiod and Simonides, after whose time it began to lose its force. Indeed, when we observe some of the remarkable variations and extremes of usage, as in Hesiod, Simonides, Empedocles, Nicander, Oppian, Tryphiodorus, we are forced to the conclusion that we are dealing, not with a process of phonetic or historical development, but with the pecu- liarities of individuals, or schools of poetry, which would prob- ably find their explanation in a detailed study of the subject- matter, style, and vocabulary of the poets in question. One question, however, may still be raised. We have exam- ined the usage of the different poets as regards correption, and have based our conclusions upon the frequency with which one or another of the vowel-endings is shortened. But are we sure that the poets exercised any choice in this matter, conscious or uncon- scious? May it not be true that they shortened all vowel-end- ings, as they came to hand, with equal freedom, so that the frequency, for example, of short at or ot, or the rarity of short y or ft), represents merely the frequency or the rarity of the occur- rence of these endings, in the language which the poets wrote ? If On Corkeption in Hiatus 249 it should appear that the results given in the table on p. 241, in regard to the relative frequency of correption in the different vowels and diphthongs, depend solely, or chiefly, upon the natural tendency of the Greek language, or of a given poet, to use words ending in these vowels or diphthongs, our whole investigation would be shown to be on the wrong path. It therefore seemed necessary to ascertain, not only the habit of each poet, in the matter of the use of the various vowel-endings, but also the usage of ordinary prose as shown in Thucydides and Xenophon. The following tables give the results. The first table indicates the number of times which the different vowel-endings appear in each poet, to each 100 lines of text (based, in most cases, on the read- ing of 600 consecutive lines), the usage of Thucydides and Xeno- phon being added below. The second table shows these same results combined into two groups as before (A— at, «, ot; B = the remaining diphthongs and long vowels), and, finally, exhibits what may be called the "tendency to shorten," which is obtained by comparing the number of occurrences of the endings in each group with the number of times they appear as short. Vowel Endings at ot ov el V 1 !)