that the articulating muscles can withstand the air-pressure as long as the glottis remains nearly closed, i.e., as long as the sounds are voiced. An increasing air-pressure will force the glottis open, thus changing the voiced stops to voiceless lenes (I, 11); these, in turn, may become fortes-aspirates- voiceless spirants (or affricates) if the force of expiration and the corresponding muscle tension continue to increase. But if the muscle tension decreases while the expiration remains normal, voiced spirants [v, o, y] will result, as is seen in the development from amado to amado, mentioned above.

(3) Voiceless spirants present a difficult problem. Obviously, increased articulation need not necessarily have any effect at all, since the breath can escape without considerable obstruction. In fact, in many cases they remain unchanged, as is seen in Germanic f, x, s under certain conditions. That part of the articulating muscles which is most inert, namely, the muscles of the back of the tongue, is apt to give way to the pressure of expiration, so that x may change to h, or even disappear (compare German hoch der höchste: höher). The change from p to t in Scandinavian (Eng. think: Swed. tänka) is due to a slight decrease of the expiration. In a number of cases, voiceless spirants become voiced. This development of [f, p, x, s] to [v, x, y, z] may be due to an unusual tension of the vocal chords (especially in the neighborhood of voiced sounds), strong enough to overcome even a high degree of expiratory force. But in general it is safer to interpret this change as a weakening of articulation a lessened force of expiration permitting the closure of the glottis; this is all the more probable since this sound transition occurs very frequently in languages of little energy of articulation — especially the change of s

to z. In certain cases, conditions are so complicated that, at this time, a definite answer seems hardly possible.

(4) While voiced spirants, too, may show a twofold development, the explanation is simple in either case. Increased expiration leads to voiceless spirants, the glottis being forced open, as in (2), while increased muscle tension, under normal conditions of expiration, changes them to voiced stops. Different conditions lead to different results, but the interpretation presents no difficulties.

(5) Liquids and Nasals are not subject to any changes of great importance, except that weakened articulation is apt to increase their vocalic (sonorous) quality — in some cases to such an extent that they become real vowels (compare especially the treatment of r in English: in a word like sir, it is either decidely vocalic, or even a vowel proper).

(6) Vowels are open sounds, and, therefore, the muscle tension is the only factor to be considered (the expiration being unhindered in the oral cavity, increased expiration will not materially affect the articulation); generally, the position of the tongue is the primary, the activity of the lips and the angle of the jaws the secondary factor.

(a) Increased articulation either causes the tongue to be raised (thru greater muscle tension), or the lips to be widened or rounded and the jaw to be lowered. Special conditions may decide for one or the other of these alternatives. In this way, [a] may change to [ox, u] in consequence of a gradual rising of the tongue (connected with the habitual lip-rounding); open [ɛ] might either become close [e], the tongue being raised, or, thru a widening of the jaw angle and a more and more energetic opening of the lips, it may change to [æ—a— AI].

Thus, the probable direction of vowel changes thru increased articulation is this:

адаю

(b) Decreased articulation must obviously lead to the opposite results, as far as it entails any change at all: [o] changes to [a], [u] to [o], and in extreme cases all vowels may result in the most decided relaxation, namely, the habitual position of the tongue when at rest ('basis of articulation,' in the narrower sense of the word), the slurred vowel [ə].

Merely as a mnemotechnic help, it may be noted that in the great majority of cases of actual occurrence, increased articulation leads to a vowel of lower pitch, decreased articulation to a vowel of higher pitch; thus the direction of change thru strengthening is: e-æ-a -o-u, while changes thru weakening show the series

NOTE. Since the basic principle of vowel change is muscle tension, while the primary cause of consonant changes is increase or decrease of the expiration, it is not surprising that languages frequently show a tendency towards weakening of consonants, but strengthening of vowels. This does not in any way contradict the principle stated in (1) Note, that contemporaneous strengthening and weakening in the same language is not to be assumed.

784213

First Cycle

The Indo-Germanic Period

BRUGMANN, Kurze vergleichende Grammatik der idg. Sprachen, 1904. OERTEL, Lectures on the Science of Language, 1901.

DELBRÜCK, Einleitung in das Studium der idg. Sprachen, 5, 1908. BLOOMFIELD, The Study of Language, 1914.

14. The Indo-European Language. Since the art of writing was not known to the Indo-Europeans before their separation, their common language has not been preserved to us in any documents whatsoever. Still, thru a systematic comparison of the individual Indo-European languages, comparative philology has succeeded in reconstructing to a considerable extent, and with a reasonable degree of certainty the original sounds and forms of the parent tongue. Strictly speaking, a 'reconstructed' language form like *kmtóm 'one hundred' means neither more nor less than a formula denoting some undefined sound combination that resulted in Lat. centum, Gk. (he)katón, Ger. hund(ert), etc., but as a matter of fact, in most cases such words are certainly not far from the actual forms used by the ancient Indo-Europeans. The sign of an asterisk before the word is used to indicate that we are dealing with a reconstructed form.

Unfortunately, there exists a great deal of confusion in the way of transcription; undoubtedly, it would be best to agree on some such system as that of the International Phonetic Association, but as long as philologists as yet have not come to such an agreement, it is better to use it only as a subsidiary means of explanation in certain cases, and in general to accept, tho in a slightly simplified form, the system in use in the majority of standard works, especially in Brugmann's Grundriss der vergleichenden Grammatik der indogermanischen Sprachen.

15. The Indo-European Consonants. According to the results of philological reconstruction, the Indo-European language before its breaking up into dialects possessed very few spirants (probably only s and z, and possibly j and w), the nasals m, n, ŋ, the liquids l, r, and a comparatively large number of stops.

According to the place of articulation, the STOPS were front, middle and back stops, i.e., labials, dentals, and velars (palato-velars); compare diagram on page 102.

According to the kinds of articulation, they were strong (voiceless fortes- either aspirated or unaspirated), middle (voiced aspirates: a glottal spirant followed the explosive), or weak (voiced lenes). The strong stops, p, t, k, are generally termed tenues, the middle stops, bh, dh, gh, mediae aspiratae, the weak stops, b, d, g, mediae.

NOTE. "Velars are sounds that are articulated at or near the dividing line between hard and soft palate" (I, 31). — For the term gutturals, see I, 31, Note.

The difference between ki-, ka-, ku-sounds (I, 31) seems to have existed to a marked degree in primitive Indo-European; originally, it was almost certainly due to the influence of the following sound (especially vowel), but in historical times analogy and other circumstances have frequently changed conditions to such an extent that a ki-sound, for instance, may also be found before an a, a kusound before an i, etc. Therefore, comparative philology recognizes three subdivisions of 'gutturals': palatals pronounced at the front palate, and transcribed by k, ĝh, ĝ; velars, pronounced at the dividing line of the hard and the soft palate, and transcribed by q, gh, g; the ku-sounds had the peculiarity that their articulation was apt to be accompanied by a simultaneous lip-rounding, so common with articulations of the back of the tongue (compare I, 41). They resembled. Engl. qu, Ital. gu, but the tongue and lip articulation took place at the same time. They are called labio-velars and transcribed by q, guh, gu. The standard transcription of Brug

mann's Grundriss reserves the signs k, gh, g for those numerous cases where it is not certain to which of the three groups a 'guttural' belongs. However, on account of the special conditions