The following investigation presents a constructional procedure segmenting an utterance in a way which correlates well with word and morpheme boundaries. The procedure requires a large set of utterances, elicited in a certain manner from an informant (or found in a very large corpus); and it requires that all the utterances be written in the same phonemic representation, determined without reference to morphemes. It then investigates a particular distributional relation among the phonemes in the utterances thus collected; and on the basis of this relation among the phonemes, it indicates particular points of segmentation within one utterance at a time. For example, in the utterance /hiyzkwikər/ He's quicker it will indicate segmentation at the points marked by dots: /hiy.z.kwik.ər/; and it will do so purely by comparing this phonemic sequence with the phonemic sequences of other utterances.

‘Homers Sprache ist nicht panhellenisch. Sie kann nur solches Sprachgut enthalten, das bei den Stämmen, die das Epos pflegten, lebendig war.’ Thus Wackernagel, SUH 222. On that principle a word limited to the Ionic dialect—'auf den ionischen Dialekt beschränkt', to use a phrase of Bechtel's—cannot gain entrance to the epos until the epos is being produced by speakers of Ionic. I present two examples: ϵ ὗντϵ, already a high-class word, doomed to early obsolescence, that could appear in poetry unchanged; and , for which an Aeolic-looking dress seemed more appropriate.

The word trík is a modern loanword in Greek, of uncertain date though probably not older than the beginning of this century. According to the definition in the Mega lexikon tēs hellēnikës glōssēs of D. Dēmētrakos (Vol. 9; Athens, 1951), this is a word of the spoken language which means ‘device, deceptive manner, artful strategem’; in short, it has some of the meanings of the English word trick. Since the last war, however, trík has been used in another sense which is fairly remote from the range of meanings just given and, as far as I know, has not yet been recorded in the dictionaries. It is employed colloquially to denote a small slip of paper bearing a printed slogan or short text suitable for distribution in the course of parades or political demonstrations. The question arises whether this word is identical with the trik just mentioned or is another word entirely.

The traditional interpretation of [y] and [w] in Spanish forms like bien ‘well’ and bueno ‘good’ has been to assign them to the phonemes /i/ and /u/. The semivowel allophone of each of these vowel phonemes is stated to occur in the following positions: SV, VSV, CSV, VSC, VS; examples (from Chavarría-Aguilar): huerta, Tehuantepec, puerta, jaula, hoy. All analysts, however, agree that there are still other allophones that cannot be assigned to /i/ and /u/, but must be assigned to a phoneme /y/ (variously transcribed as /j, y, ž/), though with the exception of King they do not describe any allophones that must be assigned to /w/ and therefore set up no such phoneme. The assignment of [y] and [w] to /i/ and /u/ therefore achieves at best only the economy that results from having no /w/ in the phoneme inventory. Other possible advantages, such as those of pattern congruity, will be examined hereafter by comparison with the analysis to be suggested here.