Abbildungen der Seite

liquids; and there is little doubt that comparative Semitic grammar in general would gain much from a thorough study of the problem.

In this connection I wish to make a suggestion concerning the nature of the initial vowel in the imperative forms of the primae n verbs in Assyrian. As is well known the 2nd. sing. imptv. of nadānu 'to give,' nazāzu 'to stand,' and naṣāru 'to guard' is idin, iziz and uşur respectively. Now we should expect forms without the n as in Hebrew ten 'give,' Syriac pel 'fall,' or with the n as in Hebrew n'fol 'fall' and in Syriac n'dor 'vow.'

The disturbing initial vowel which is found instead in Assyrian has been explained as the result of analogy with the dissylabic imperative forms of the 'kušuď' type.* In view of the irregular treatment which in the various Semitic dialects is accorded to the n in the above-cited position I would suggest that we have here different modes of approach to an initial anteconsonantal nasal that tends to become vocalic because of this very position. In Hebrew as well as in Aramaic this nasal did not develop into a vowel and was, in consequence, either entirely dropped (ten, pel), or was restored to full consonantal value with the aid of a following Shwa (nofol, nođor). In Assyrian, on the other hand, a vowel developed whose character was determined by the vowel of the preterite stem of the verb in question. That vowel, therefore, is not a heterogeneous importation caused by analogy, but, to my thinking, the phonetic heir of the original nasal."

The same explanation holds true for forms like itaşru (for nitașru) from ittaşar <intaşar (√naşaru 'to guard') and itpuşu for nitpușu (√napaşu 'to dash'). In the above examples there cannot be even the slightest suspicion of analogical extension as the number of syllables remains the same in either case. But a regular development of an initial anteconsonantal n into a vowel would easily account for the otherwise unexplained representations.

* See Brockelmann, Grundriss der Vergleichenden Grammatik der Semitishen Sprachen, 59 ff.

To follow Professor Haupt in assuming that forms like idin developed out of the preterit form 'by a simple dropping of the initial n' (a view expressed at the last meeting of the American Oriental Society, New Haven. 1925), is to ignore the fact that the preterite of nadānu is made from *indin and not from *inidin.




In my Studies in New Mexican Spanish1 I have already called attention to the existence of certain syllabic consonants in New Mexican Spanish. In the following pages I shall discuss this subject much more in detail.

It is well known to students of language that syllabic consonants are to be found in some languages. A regular series of syllabic consonants, properly classified and well understood with respect to their phonetic developments in the various related languages, has been established for Indo-European.

In the Romance languages there are no syllabic consonants in standard speech. To my knowledge none are found in any of the Romance dialects with the single exception of New Mexican Spanish. Our present study of the New Mexican Spanish syllabic consonants offers, therefore, some new and interesting material that may give us new light on the development of syllabic consonants in Indo-European and Latin. For that reason, although the present study is in general only a collection of material, references are given to parallel developments in the Italic languages.3

The New Mexican Spanish syllabic consonants are the following: A. Long voiced consonants

[merged small][merged small][ocr errors][merged small][merged small][merged small]

1 Part I: Phonology (Revue de Dialectologie Romane, 1909), §167.

There are, of course, a few sporadic examples of syllabic consonants in standard French and Spanish in certain exclamations, such as French chut, pst, Spanish pst, Italian sst, and perhaps in a few of the monosyllabic expressions of approval or disapproval, etc., that are commonly heard in many languages but seldom recorded graphically.

In the study of comparative developments with Latin and other Italic languages I shall refer principally to Sommer, Handbuch der lateinischen Laut-und Formenlehre, Heidelberg, 1902, and Lindsay, The Latin Language, Oxford, 1894.

The basis for the above classification is involved in the description and discussion of the sounds that follow. The five phonetic symbols above given represent the five New Mexican Spanish syllabic consonants commonly heard in New Mexico and Southern Colorado."


I shall now give a detailed account of the exact phonetic values of the above five phonetic symbols that represent the five New Mexican Spanish syllabic consonants."


Long and short consonants. A classification according to long and short may appear somewhat arbitrary and borrowed from the IndoEuropean system. It is, of course, a somewhat different problem. In the New Mexican Spanish syllabic consonants no exact or nearly exact quantitative values are involved. There is, however, a very great quantitative difference between m, n, 1, and Į, ǹ, that has a great importance upon articulation. The voiced syllabic consonants, m, n, Į, are very long and may be held indefinitely. Being strongly accented they are usually as long as the consonant plus the accented vowel absorbed. The other two, however, the I and the r, are very short in duration and there is a tendency towards a rather loose and indistinct consonantal articulation.

General characteristics. All the New Mexican Spanish syllabic consonants are voiced. Their vowel quality does not seem to be uniform. In syllabic m it is a sound between the English u in mum and the French e in le. In n,,,, it is a sound similar to French e in le. The articulation consists of the vibration of the vocal chords and the consonant-vowel quality determined by the organs of speech above the larynx. In the

A map of the New Mexican Spanish territory covered by my investigations into New Mexican Spanish language and folklore is given in my Studies in New Mexican Spanish, Phonology, page 6. In general it may be said to cover all of New Mexico north of Socorro and all of Colorado south of Saguache and Pueblo.


For the symbols of the syllabic consonants I employ those used by Sommer, but their phonetic values are, of course, the ones established for New Mexico. The acute accent over the syllabic consonant is used to designate a long syllabic consonant and with stress accent, the grave accent to designate a short one with stress accent. The absence of both means a very short syllabic consonant and with no stress accent. In the phonetic transcriptions of other consonants (not the syllabic) and of the vowels, I employ the phonetic alphabet used by Tomás Navarro Tomás in his Manual de Pronunciación Española, 1st edition, Madrid, 1918, 2nd edition, 1921.


* As is well known, Spanish has a very pronounced intensity stress or accent. The syllabic consonants m, n, I, have the same intensity stress as the accented vowels they absorb, and quantitatively they may be even longer. In the Germanic languages syllabic consonants are usually not accented.

case of m, n, I, I there is no consonantal friction in the syllabic consonant proper, but only a consonantal vowel, the quality of which is determined by the firm and fixed position of the lips or tongue while the breath passes through the nasal or oral cavities or both. In the case of the there is also consonantal friction during the articulation of the syllabic consonant proper.

Aside from the vibration of the vocal chords or voice, the syllabic consonant consists essentially of the fixed position of the lips in m (completely closed), and of the tongue-teeth and tongue-alveoli contact, firmly held in ǹ, I, and I, partially closed in . The syllabic consonant r. lasts as long as the vibration of the vocal chords plus this complete and firm or partially closed contact continue. The syllabic consonant proper consists, therefore, of voice plus the attack or obstruction. When the obstruction is removed (the lips opened in the case of m, or the tongue removed in the case of n, I, I,) we have what we might call the consonantal release, and this release is a new consonant and begins a new syllable. The attack begins with a marked rise in intensity and a new breath impulse and then the intensity gradually diminishes. But none of the New Mexican Spanish syllabic consonants occur at the end of a phonic group. The release of the obstruction becomes necessary, and this release of the attack or contact which is marked by a new rise in intensity becomes the consonantal articulation that begins the new syllable. Before a vowel, the release or new consonantal articulation is determined by the previous consonantal attack; n becomes nn. Before a consonant that is articulated by the organs of speech in the position of the previous consonantal attack, the release becomes that consonant. And before a consonant not previously anticipated by the position of the organs involved in its articulation, the release takes place simultaneously with the articulation of the new consonant.

To describe their consonantal articulation, each syllabic consonant has to be treated separately.

1. Long voiced nasal m. This syllabic consonant is a voiced, bilabial nasal, articulated with the lips completely closed. The breath passes through the nose, and aside from its syllabic character and the fact that the lips remain closed, its general character is like the ordinary Spanish m described by Navarro Tomás. As we shall see later, syllabic m often attracts a following ordinary consonantal m that becomes the consonantal release of the syllabic m, the result being mm. This release, however, may actually take the form of any labial consonant that follows.

↑ Manual de Pronunciación Española, §87.

2. Long voiced nasal n. This syllabic consonant is ordinarily a voiced infra-alveolar nasal. The blade of the tongue is pressed against the lower surface of the alveoles, the tip sometimes touching the inner surfaces of the upper incisors. See Fig. 1. In the San Luis Valley, Southern Colorado, however, it is usually a wide denti-alveolar nasal, the front of the tongue being pressed against the upper teeth and alveoles, and the tip protruding to an almost interdental position. See Fig. 2. When it falls before a vowel the release or tongue withdrawal articulates the consonant n. Syllabic n then becomes nn. And when pronounced immedi

Fig. 1.

Fig. 2.

ately before dentals or alveolars the release becomes the corresponding dental or alveolar articulation with the necessary assimilation to the nasal n.

3. Long syllabic I. Syllabic I does not seem to have in the New Mexican territory a very uniform articulation. The most common articulation is that of a voiced, denti-alveolar lateral. The tip of the tongue is pressed against the inner surfaces of the upper incisors, and the blade of the tongue is pressed against the alveoles over a wide area. See Fig. 3. Syllabic may attract a following 1, the result being II, with only one consonantal release.

4. Short syllabic I. There is also heard in New Mexican Spanish a short, weak, unaccented syllabic I in the initial groups, cl-, gl-, when


« ZurückWeiter »