ALPHABET. The name alphabet is derived from the names of the first two letters of the Greek alphabet, alpha and beta, and denotes a set of characters, or, as we call them, letters, each of which represents a given sound or sounds. This representation is necessarily rough and of quite a general character. This is easily seen in the case of the first letter of the English alphabet, for example, which represents different sounds in the words father, man and take. But even in cases where a letter is regard ed as representing a single sound, it does so roughly, taking no account of differences of intonation, tone or pitch, nor of stress, nor of slight variations of pronunciation which vary not only between one individual speaker and another, but also from time to time in the case of an individual in accordance with the posi tion of a given sound in a word, of a word in a phrase, or with the nature of the phrase to which he is giving utterance. In this connection writing stands in much the same relationship to speech as speech does to thought. If language is not a sufficiently deli cate instrument to express the nuances of human thought, writing is a less delicate instrument still, and any attempt to multiply signs and characters to keep pace with the subtle variations of the human voice would only impair their usefulness.
An alphabet is a highly-developed, artificial form of writing. The connection between sound and character is conventional and not essential. This is not the case with all forms. Pictographs (see PICTOGRAPHY), ideographs (q.v.), and hieroglyphs (q.v.) are forms of writing the characters of which bear an essential rela tionship to what they are intended to represent. Such methods constitute generally speaking an earlier stage in the development of writing than syllabaries and alphabets, which have passed into the conventionalized stage. An earlier stage still may be found in the various devices employed to assist memory, such as the notched stick, notched in the presence of the messenger to whom the significance of each notch is verbally emphasized, a method not far removed from the tying of a knot in the handkerchief (see KNoTs). Again there are the Peruvian Quipus (q.v.), notched sticks from which strings are suspended of varying colour and length, knotted in various places, each variation having its own significance; or the North American Indian custom of adorning deerskin belts with wampum beads (see WAMPUM), whose colour and position were employed to describe historical events such as battles, or even the terms of treaties, with remarkable accu racy to those who knew the secret of their elucidation. There are thus three distinct stages in the growth of writing, though they need not necessarily be thought of as hard and fast chronological stages.
Picture-writing or the use of the ideograph is the intermediate of the three stages. An ideograph (q.v.) is a drawing representing not a sound, nor even a word, but an idea. Thus the drawing of a man with protruding ribs represents famine ; or an eye with tears dropping from it, sorrow. Till recently such methods of communication were in use among certain tribes of North Ameri can Indians. They appear to have been so among many races in various regions.
In this connection should be mentioned the writing known as Nsibidi or Nchibiddi in use among a secret society or club among the Ekoi people of southern Nigeria. The writing is purely ideo graphic, but the symbols are conventional, fixed in form and meaning. The script is jealously guarded by the members of the secret society. Its origin is unknown; but certain resemblances have been detected to the Egyptian hieroglyphic system, one of the most remarkable being that the Nsibidi picture-symbol for house is rectangular in shape, whereas all dwellings among the peo ple who use it to-day are, and have been for centuries, round. Moreover the Ekoi people are semi-Bantu in descent, and there is reason to suppose that the Bantu peoples migrated from the lower Nile regions. (For an account of this writing see In the Shadow of the Bush by P. Amaury Talbot, Appendix G. P. 447 ff.) Six different peoples in various parts of the world developed a system of writing usually known as transitional, that is to say, they continued the use of the ideograph, but added symbols of a purely phonetic value, and made use of the two side by side. These peoples were the Sumerians and Babylonians, whose cunei form was a script of this nature, the Egyptians, the Hittites, the Chinese in the Far East, and in America the Mayas and Aztecs.
The method by which this change took place was, generally speak ing, that the picture-symbol representing any given object, an eye, for example, came to represent the same sound under all circumstances, as if for instance the drawing of an eye came to stand not only for the meaning eye, but also for the pronoun I, which has the same phonetic value. This could be the case not only when the sound constituted a whole word, but when it occurred as a single syllable within a word. Even in purely pictographical writing something of the sort occasionally took place. The Mexican name itz-coatl meaning literally snake-knife and represented by the drawing of a snake surrounded by knives, occasionally occurs as itz-co-atl with a different picture, co mean ing a pot. Though this process is of a different nature, it serves as an illustration of the tendency to connect the drawing not only with the object it essentially represented but also with any homonym of that object. This being so, it became the custom to draw or write the picture with greater haste and carelessness, time being more important than accuracy, so soon as the symbol came to represent not an object but a sound. A badly-drawn eye, for will serve perfectly well for a sound that covers both words eye and I. There is no longer need for the drawing to be recognizable as an eye when it is no longer required to repre sent the object to the sight, but only a sound which itself carries meaning to the ear. Thus picture-writing degenerated, the process being a distinct gain in ease of communication. It was in this way that the Egyptian hieratic and demotic scripts came into existence, although so strong was the habit of writing hieroglyphs that for many cent' ries both were employed side by side, even literally so in the same piece of writing, the hieroglyph sometimes actually following the phonetic symbol in order, as it were, that the meaning mignt be doubly assured.
The Chinese ideographic system exhibits peculiarities which depend upon the language that it was formed to represent. So soon as the symbols attained a phonetic value, serious difficulty arose from the fact that a single monosyllabic word might have many distinct meanings, distinguished in speech only by intona tion which it was, for a time, impossible to represent in writing. The difficulty was felt in the spoken language also, and one homonym was distinguished from another by the addition of a second word which made the meaning clear, a kind of compound being thus formed. In the course of time the original pictographs have deteriorated, but they are capable of analysis into their original constituent parts without undue difficulty (see CHINESE LANGUAGE) .
The Japanese, when they came into contact with Chinese civilization in the 5th century A.D., adopted the Chinese phono graphic script and endeavoured to adapt it to their language. Finding, however, that their chief sounds could be comprehended in a syllabary of 47 characters they developed these from Chinese characters. For literary or learned purposes the Chinese phono grams are still in use among them, so that in spite of the utter difference of language the Japanese and Chinese can communicate by means of writing, although each symbol has a totally different phonetic value to the speakers of either language. The neighbour ing countries of China, Japan and Korea exhibit three separate modes of writing, ideographic, syllabic and alphabetical, the Koreans having invented an alphabet based on Sanskrit forms.
The syllabary and alphabet are examples of the most highly developed stage of writing, the conventional stage, when there is no essential connection between sound and symbol. To some peoples the development of a syllabary seemed to come more easily than that of an alphabet. A syllabary existed in Cyprus .in ancient times. In the case of a language that for reasons of phonetic decay or otherwise has multiplied consonants in a single syllable the syllabary is a cumbersome mode of writing. The word strength, for example, would have to be written as-at-ar-eng-ath, or sa-ta-re-nga-tha and this is obviously a repre sentation of the sounds composing the word that is far from adequate. It will also be readily seen that a much greater number of symbols is required for a syllabary than is the case with the far more convenient alphabet.
The alphabet then is the form of writing that to those peoples who have developed, borrowed or adopted it has been found the most convenient and adaptable. Its use is acquired in childhood with ease, which is far from being the case with the Chinese phonograms, for instance. It may also be passed from one language to another without difficulty. The history of our own alphabet, which has survived as an alphabet with, all things con sidered, surprisingly little change for nearly 3,00o years and is still vigorous, untouched by the introduction of printing or of the typing machine, demonstrates its suitability to the needs of the many languages it has served.
The story of our alphabet from the time that we first find it employed in the earliest known Greek inscriptions is not hard to trace. It is its history in its pre-Greek days that is still wrapped in doubt. The Greek names for its letters, alpha, beta, gamma, delta, etc., demonstrate its .Semitic origin, for though the names may not in every case have a Semitic etymology, they correspond closely with the Semitic names (cf. the Hebrew alepli, beth, gimel, daleth). Two questions have hitherto remained un answered. How did the Greeks obtain it from the Semites? And what was its pre-Semite history? In recent years decided advances have been made towards a solution of either question.
Until the second decade of the loth century the earliest known examples of the Semitic alphabet were what is known as the Moabite Stone, an inscription of Mesha king of Moab dating from the 9th century B.C., and a votive inscription to Baal-Lebanon discovered in Cyprus. Various theories have been current as to the origin of the Semitic alphabet, the earliest being the view of Lenormant, published by De Rouge in 1874, that Egypt was its starting-place. Others attempted to connect it with Babylonian cuneiform, with the Cyprian syllabary, or with the Minoan writ ing of Crete. The latter is the view of Sir Arthur Evans who, in his great work Scripta Minoa, develops the theory that the alpha bet was taken from Crete to Palestine by the Philistines and from them borrowed by the Phoenicians. The Egyptian view was revived in 1916 in a paper published in the Journal of Egyptian Archaeology, vol. iii. (1916) p. 1 ff., by Dr. Alan Gardiner. This paper deals with certain inscriptions discovered in 1906 by Sir W. Flinders Petrie in the Sinaitic peninsula, in which Dr. Gar diner sees a stage of writing intermediate between Egyptian hieroglyphic and the Semitic alphabet. The script appears to be fully alphabetic, and the date of the inscriptions is not later than 1 soo B.C. These inscriptions have also been dealt with by Hans Bauer (Zur Entzi f erung der neuentdeckten Sinaischri f t, Halle, 1918), who is inclined to see in the script a prototype of the Semitic alphabets, independent of Egypt.
For a solution of the question the evidence of the four south Semitic alphabets must be considered. These are known as Sabaean, Lihyanic, Thamudenic and Safahitic. Inscriptions in these south Semitic alphabets were discovered in considerable numbers during the course of the 19th century. Those in the Sabaean alphabet come from the district of Yemen in south Arabia, the Lihyanic from El-Ola in the north of the Hejaz.
The Safahitic inscriptions come from rocks in the district of Safah near Damascus. Most of the south Semitic inscriptions read from right to left but a few are written f3ovrrpocka6v, as the Greeks said, that is to say, from right to left and 'left to right in alternate lines. There are 29 letters in the Sabaean alpha bet. The date of the earliest Sabaean inscriptions appears to be about the 6th century B.C., and most of those who have noticed them are agreed that this alphabet descends not from that of the Moabite Stone but from an older source common to both. Dr. Alan Gardiner sees this source in the alphabet of the Sinai inscriptions. Two or three of the signs, notably that for beth [ in Sabaean, for phe 0 in Sabaean, and for thau — in Thamu denic and Safahitic resemble symbols of the Sinai script. But it would be difficult to derive most of the signs from any found there. Further, Dr. Gardiner's identification of the Sinai symbols does not rest upon a completely sure basis in the absence of a certain interpretation of the inscriptions. Dr. Gardiner sees the most important line of evidence in the names of the letters. Of the 2 2 letters of the Semitic alphabet 17 have Semitic names, which are the names of ordinary objects. Five are unintelligible. It is obvious that the simplest interpretation of the fact of these names is that the letters were once picture-symbols representing the object whose name the alphabetical symbol has retained. In 1 s cases symbols appear on the Sinai inscriptions which evidently represent objects to which these names belong. It is therefore no far-fetched conclusion to regard these symbols as prototypes of the alphabetical symbols which continued to bear the names of the objects they represent. There is also no doubt of the close connection of the symbols of the Sinai script with certain Egyptian hieroglyphics, and it is scarcely probable that identical picture symbols would have developed independently in Egypt and so near to its borders as to be in use in the Sinai peninsula. The close intercourse between Egypt and regions lying beyond Sinai render complete independence of script unlikely. On the other hand, of 32 separate characters apparently employed in the Sinai inscriptions, 17 are unintelligible, not being found to cor respond sufficiently closely either with Egyptian hieroglyphic characters on the one hand nor letters of the Semitic alphabets on the other. The accompanying table, taken from that drawn up by Dr. Alan Gardiner to illustrate his article shows the rela tionship between the Egyptian hieroglyphs, the Sinai script, the alphabet of the Moabite Stone, the south Semitic alphabets, and that of early Greece. While Dr. Gardiner's interpretation of this script cannot be said to have proved that Egypt was the original home of the Semitic alphabets, it has directed attention to Egypt as deserving serious consideration as a claimant to this honour.
A different view of the origin of both the Phoenician and Greek alphabets is offered by Prof. Sir W. Flinders Petrie who in 1912 published his book The Formation of the Alphabet. In this he argues that both these alphabets together with those of Asia Minor, the south Semitic, Cyprian and certain Egyptian scripts developed from a series of conventional marks or signs employed for commercial purposes throughout the Mediterranean area from earliest times. "In Egypt, especially," he says, "the monumental evidence shows two entirely different sources of conventional marks. In late times the picture writing passed through many stages, until it became the complex grouping of slightly varying strokes in the demotic writing. But far before all this there had existed, from the beginning of the prehistoric ages, a totally different system of linear signs, full of variety and distinction. This early system was certainly in its decadence long before any hieroglyphs were used in Egypt. Similarly in Crete a system of linear signs precedes the pictographic records." But Professor Petrie is alone, or practically so, in supposing that the early marks to which he refers had any significance, and his theory of the development of various local alphabets from such marks has not found general acceptance. Too many links are missing in the chain of proof. The fact that identical alphabetical symbols occur in widely distant regions of the Mediterranean area which he stresses, is capable of other explanations.
The most important fact ignored by any theory that would derive the Greek and Phoenician alphabets independently from a source older than either is that the names of the letters, as far as they have a known etymology or meaning, are Semitic. He brew aleph, beth, gimel, daletli, etc., correspond unmistakably with Greek alpha, beta, gamma, delta. The Semitic names are words in the Semitic languages. The Greek names are meaningless in Greek. It has indeed been argued that the Greek names were the original ones, meaningless, and taken in the first instance simply from the sounds represented by the letters ; that the Phoenicians took over the alphabet together with the names from the Greeks and adapted the names by the process known as f olk etymology. But the Sinai inscriptions, discussed above, may be said to have rendered this theory untenable. If they do not repre sent a link between the hieroglyphic system of Egypt and the Phoenician alphabet, they represent a prototype of the latter, appearing at a date and at a place that render any theory of bor rowing from Greek sources highly improbable. It is reasonably certain that the Greeks derived their alphabet from Semitic sources, taking over the names with the letters. The names are not identical. The Greek names end with a vowel—alpha for aleph, beta for beth, etc. The vocalization has been explained merely as being more in agreement with Greek morphology, but this is not a very definite or satisfactory explanation. Isaac Taylor had already pointed out in the History of the Alphabet (1883) vol. ii. p. 27 that the Greek forms are taken from Aramaic, Aramaic showing an emphatic form of the root which ends in a vowel and drops the vowel in the preceding syllable. He postulated two separate borrowings of the alphabet by Greeks, the earlier being the Chalcidic alphabet borrowed with the Baby lonian system of weights and coinage from Aramaic peoples in the south-eastern extremity of Asia Minor, possibly through Lydia ; the second the borrowing by the Ionians directly from Phoenician traders. The Aramaic names came with the first borrowing and were carried on and applied by the later borrow ers. These two forms of the alphabet, Chalcidic and Ionic, gen erally known as western and eastern, represent two different bor rowings, but we may probably rule out the theory of any direct borrowing from Phoenician sources. The tradition that this was so comes from Herodotus, who refers only to Boeotia, and it is probable that the name Phoenices was a vague term for eastern foreigners. The Achaeans were a naval power settled in the east ern Aegaean in the isth century B.c. There is therefore no difficulty in concluding that they borrowed the alphabet from peoples with whom they were in close contact in western and southern Asia Minor at a considerably earlier date than has hitherto been supposed; nor that, independent as Greek States were, there were separate borrowings and separate adaptations.
The chief adaptations made by the Greeks were the allocation of certain of the 22 Semitic letters to vowel-sounds, and the addition of certain letters. The oldest extant Greek inscriptions are those found in 1896 in the island of Thera, which date from the 8th, possibly even the 9th, century B.c., the inscriptions from Abu Simbel in Upper Egypt, dating from the 7th century, a Corinthian and an Athenian inscription of the 6th century. In Thera the oldest inscriptions are written from right to left, others are ovarpo n b6v, others still later read from left to right. In the Lydian alphabet, known to us by inscriptions found at Sardis in the beginning of the present century, these symbols also represent vowels. Most Greek States developed local variations either in the forms of certain letters, or in the representation of sounds not represented by any of the 22 letters taken over. The two main divisions, eastern and western, however, remain till the 4th cen tury B.C., when in the main uniformity of alphabet was attained .throughout Greece. The early inscriptions from Thera show a remarkably close resemblance to the forms of the Semitic alphabet as represented on the Moabite Stone or the Baal-Lebanon inscrip tion from Cyprus. The alphabet of Abu-Simbel is of the Ionic type.
The two main divisions, eastern and western, again sub-divide into two, the cities of Asia Minor and the adjacent islands with their colonies using forms showing in some cases slight differences from those in use in the Cyclades and Attica. Again the alphabets of Megara, Corinth, Sicyon, Phlius and western Argolis with their colonies differ from the west which includes Euboea, Boeotia, Phocis, Locris, Thessaly and the rest of the Peloponnese. Through out all Greek alphabets the Semitic letters aleph, he and ayin, which represented breathings, were adapted to represent vowels. Alpha (aleph) was consistent throughout. The letter correspond ing to Semitic he (later epsilon) represented both the short and long vowel e in those cases where eta represented the rough breathing. Generally speaking eta represented the vowel-sound in the eastern alphabet, the breathing in the western. Thus in the west the symbol or E is used for the two sounds represented in Attic by e and rt ; in the east it more usually represents the short vowel alone. In addition to this, however, it is occasionally used throughout to represent the long close a (the long a repre sented by eta was open), which in Attic is written as if it were a diphthong, et. The Corinthian alphabet is peculiar in using (up to the 6th century and later in outlying regions) the sign g or B to represent e, the long close vowel being written B5. Sicyon uses a symbol with the f orm X. Semitic cheth, Greek eta, was used, as we have seen, in the east generally speaking as a vowel, in the west as the rough breathing. In the Theran inscriptions, it occurs in both capacities. The semi-vowel (resembling the sound of English initial y) had disappeared from the Greek language in prehistoric times, degenerating into a rough breathing, so that Semitic yod was adapted in Greek to represent the vowel c (iota). Koppa (9) had disappeared in the eastern alphabets by the Sth century B.C., its place being usurped by kappa ( K ), but it lingered in the west and was introduced into the Italic alphabets, as we shall see.
The origin of the digamma (F, representing the w sound) is obscure. Vau, the sixth letter of the Semitic alphabet, was taken by the Greeks for use as a vowel, exactly as they had taken yod, and placed at the end of the alphabet following tau. But whereas the consonantal i sound did not exist in Greek, the consonantal u-sound (akin to w) survived till classical times in certain dialects and needed a symbol to represent it. Thus the digamma took the place of van in the order of the alphabet, but opinions differ as to whence it derived its form. Some hold that it was differentiated from the preceding letter by the omission of one of the horizontal strokes, others that it descends from vau (y). The latter view is rendered less difficult by the fact that N! occurs as a form of the digamma from Oaxos in Crete. In the absence of further evidence it is impossible to reach any definite conclusion.
The treatment of the sibilants of the Semitic alphabet by the Greeks is complicated. Zain was taken over as zeta, though its pronunciation in early times is not certain and may have varied throughout the dialects between dz, dz, z and z. The zsth letter of the Semitic alphabet was Samech the name of which through its Aramaic form simcha became in Greek by metathesis sigma. The name, however, was transposed to the 21st letter, which in Semitic was saga. The letter itself was retained in the eastern or Ionic alphabet with the value ks (i). In Thera, however, in early times it has the value of ss, x or ks, being expressed by KM.
The usual manner of expressing this combination in the Aegaean islands was X5 . The letter X, which often appears in the west as +, may be a direct descendant of (samech), and quite independent of X or + , the 26th letter of the Greek alphabet, the origin of which is unknown. The sibilant s was expressed by two separate symbols, M and 5 or Z , the one descended from Semitic ssade, the other from Semitic san. Both do not appear in the same alphabet. M appears in Crete, Thera and Melos, in Phocis in the 6th century, elsewhere only in the Peloponnese and its colonies. In two 5th century inscriptions from Teos and Halicarnassus a form T appears with the value ocr. The un voiced velar aspirate was expressed by a form X or -1-, which became the 26th letter of the alphabet in practically all alphabets except in the western group. The same form was used in the west for the x-sound, and here it was probably a descend ant of Semitic samech . In the west where X was used for the x-sound and not for the unvoiced velar aspirate (kh, X), the latter was expressed by a symbol . In Boeotia we actually find V used in combination with .5 to represent the x-sound, a fact that suggests that there was some consciousness that it was used in the west as a substitute for eastern X. is used singly to express the x-sound in Thera and Melos. In the east, where X expressed the kla-sound, 41/41j was used (sometimes with the form 4) to express the combination ps, a use in which it was later standardized by the spread of the Ionic alphabet.
Other symbols not derived from Semitic were (i.) 1 (phi) used to express the unvoiced labial aspirate (ph) and (ii.) fi the last letter of the Greek alphabet, probably differentiated from 0. In the west this symbol appears very seldom. In the east it represents the long open o. In the Cyclades curiously enough it frequently expresses the close vowel, whether long or short (o or on) while 0 represents the open sound.
The origin of the symbols 1 and 4/ is unknown. In the case of X or 4 it is probable that we should differentiate X derived from and standing for x from X standing for kh. The origin of the latter is however obscure. It is possible that these signs were adapted in Asia Minor from the Cypriote syllabary.
Various local alphabets were in use in Asia Minor that were parallel to the Greek. The Lydian alphabet for instance is known to us from inscriptions dating from the 4th century B.C. A few of these were discovered in the last century, 36 were found by an American expedition in Sardis between the years 1910 and 1913, the rest by J. Keil and von Premerstein in various parts of ancient Lydia between 1906 and 1912. The alphabet consisted of 26 symbols. The parallelism with the Greek and Semitic alphabets is clear, but there are additional symbols the sound of some of which is uncertain in the absence of a completely certain interpre tation of the language of the inscriptions. The most interesting fact about this alphabet is the identity of the symbol for f 8 with the Etruscan symbol for the same sound.
The western Greek or Chalcidian alphabet was brought to Italian soil by the Greek colonies of Magna Graecia, and until recently it was confidently held that the Latin alphabet, the parent of our own, was derived from this Chalcidic alphabet by contact with the Greek colonies of southern Italy. Too little was made of the influence of Etruria because too little was known of that country and its people. Six Italic alphabets have come down to us. These are the Latin, Faliscan, Etruscan, Umbrian, Oscan and Messapian. The Messapian is purely Greek in origin and was used by a people who were isolated in Calabria and spoke an Illyrian dialect. The Umbrian and Oscan alphabets are derived from the Etruscan. The district in which the Umbrian alphabet was used was east of the Apennines. The Oscan was in Campania. Thus the Latin and Faliscan alphabets were bounded on all three sides by the Etruscan and alphabets derived from it. In these circumstances Etruscan must have exercised a strong influence upon the Latin alphabet, following the impact of Etruscan civiliza tion upon Roman in other directions.
The Etruscan alphabet is known to us from certain tomb in scriptions, from part of an Etruscan book that was used as the wrapping of a mummy, and from three alphabets or abcdaria found at Marsiliana di Albegna (near Siena), at Formello (near the site of the ancient Veii), and at Cerveteri, the ancient Caere. It has F (digamma) as f, and retains forms corresponding to all three Phoenician sibilants, samech, san and ssade. It has X repre senting ks, for ph and 3' as ch. On one tomb inscription of the 8th century B.C. 8, is found representing f, identical with the same character used for the same sound in the Lydian alpha bet (see above). The view now usually held as to the origin of the Etruscan alphabet is that it was brought by its users to Italy when they migrated thither from Asia Minor probably in the 9th century B.c., and that they had received it while still in Asia Minor from Greek sources at a period prior to the division of the Greek alphabet into eastern and western. Unfortunately we do not know the Etruscan names for the letters of the alphabet. There is nothing impossible, nor even improbable, in the view of the derivation of the Etruscan alphabet from Greek sources in Asia Minor at an early date, since the Achaeans were of con siderable importance in the eastern Aegaean as early as the z 5th century B.c. It may be taken as reasonably certain that the Etruscans did acquire their alphabet in Asia Minor. One impor tant point of contact with the Lydian alphabet has been already noticed, but although there was a close mutual relationship be tween the Greek and other alphabets in Asia Minor in the first half of the first millennium B.C., the evidence does not yet indi cate with precision who were borrowers and who lenders.
The oldest extant records of the Latin alphabet are firstly the inscription known as the Dvenos Inscription (it reads D V E N OS MED FE C E D) found in r 88o in Rome upon an earthenware vessel with three separate branches, dating from the former half of the 4th century B.C. In the second place is the Praeneste fibula, dating from the 6th or 5th century, the inscription upon which runs from right to left. It reads MANIOS MED FHEFHAKED N U M AS I O L, and is remarkable as illustrating the device of combining the letters F (digamma) and H (aspirate) to repre sent the sound of f which was common in Latin, but did not occur in Greek. In the Dvenos Inscription F alone is used, a develop ment that may be due to Etruscan influence ; for the Romans had the w-sound in their language, but instead of using F to repre sent it, they rather clumsily allowed V to represent both this and the vowel-sound u. The third instance of early Latin is an inscrip tion upon a column, written (3ovaTpo4n&6v found in the Roman forum and dating probably from the 5th century. In it K is in ordinary use as a consonant, and the word R E C E I occurs where C has still the force of g.
The third letter of the alphabet C, which represented in Greek the voiced velar stop (the sound of g), came in Latin to represent the unvoiced stop. The reason for this is obscure, but Etruscan influence cannot have been absent, for the voiced stops were un represented in the Etruscan language. Latin did not lose the voiced stop but adapted a symbol G from C to represent it, which it placed in the position of = (zeta) for which it had no use. C thus took the place of K, which fell out of ordinary use, remaining only as the initial of well-known or official words. C also remained with the force of g in the initials of proper names. F (digamma) was used at first in combination with H to repre sent the f-sound. Later the H was dropped, again probably under Etruscan influence. When after the conquest of Greece Greek words began to be borrowed by the Latin language, the symbols Y and Z were adapted from Greek for use in such words and placed at the end of the alphabet. There were no sounds in Latin to correspond to the three aspirates. Q (theta) was retained to .represent the numeral roo, and became modified to C and identi fied with the initial letter of the number. ® (phi) written C ID was used to represent i,000 and became identified with M, the initial letter of mille. l (clii) became then L, and was used for 5o.
The Latin alphabet of 23 letters (including Y and Z became extended during the course of the middle ages to our own of 26 by the division of I into i and j and the tripartite division of V into u, v and w. I developed in the i 5th century as an initial (more or nate) form of i, and as the consonantal sound occurred more usually at the beginning of words and the vocalic in the middle, j became specialized to represent the consonant, i the vowel. The history of u and v is precisely similar except that it took place five centuries earlier. W arose, as its name implies, out of a com bination uu or vv, which about the i i th century came to represent in old English the w-sound which had previously been repre sented by a rune.
It only remains to consider briefly certain alphabets which derived from the same source as our own. The alphabets of India are offshoots of the stem. All derive from two sources, Kharosthi and Brahmi. Though attempts have been made to show that the Brahmi alphabet was developed in India itself without connection with outside sources, this contention cannot be maintained. It has a manifest kinship with the Semitic alphabets. Its origin is obscure; but there seems little doubt that it derives from the south Semitic group of alphabets through contact with Sabataean traders. The Brahmi alphabet is the parent of all modern Indian alphabets. Kharosthi is an adaptation to the needs of the Indian languages of the Aramaic alphabet, which was in use in north west India in the days of Persian rule. Inscriptions in it first appear in the 3rd century B.C. None have been found of a date subsequent to the 5th century A.D. By the 3rd century A.D. it had spread to Chinese Turkestan. It was superseded in all districts by Brahmi.
The alphabets in use in Persia, at least from the time of the Arsacid dynasty onwards, are based upon the Aramaic. Cuneiform was the writing of the Achaemenid empire, but after the Greek conquest Aramaic was introduced and became general. The Pahlavi is the alphabet of the Sassanid kings. The Arsacid and Sassanian alphabets are found together in the liturgical inscrip tion of Hadji-abad, as well as in the Paikuli inscription dating from about 30o B.C. Ernst Herzfeld dates the commencement of Pahlavi in the time of Darius the Great (vd. Paikuli p. 73).
Modern Arabic writing has developed from a form of Aramaic used in the land of the Nabataeans, an Arabian people. The oldest record of it now extant is a stele from Tema, in north-west Arabia, dating from the 5th century B.C. or perhaps earlier. Later inscriptions, known as Nabataean, belong to the period between 9 B.C. and A.D. 75. The graffiti on the rocks of Mount Sinai carry the record of the development of this alphabet down to the 3rd century A.D. In the early Mohammedan period two types of Arabic writing existed, known as the Cufic and the nashki. The former was discontinued except for formal purposes, where cur sive writing could not be employed. The nashki is the parent of modern Arabic writing, in which many characters have become so similar owing to the degeneration of the cursive script that it is necessary to distinguish them by diacritical points. The Chris tians of northern Syria employed a form of Arabic alphabet, known as Estrangelo, and this was taken by Nestorian missionaries into central Asia, becoming the ancestor of a great number of alphabets which spread as far east as Manchuria.
The earliest records of Aramaic go back to about Boa B.C. and were found at Sindjirli in northern Syria. The alphabet at this time differs little from that of the Moabite Stone. Aramaic became by far the most important and widespread of the north Semitic alphabets. It was used in Assyria as a cursive script side by side with cuneiform. It is marked by two distinct ten dencies : (i.) the opening of the heads of the letters, beth (9) becoming V , daleth (A) LI and resh (4) y. 0 became U, then V ; (ii.) angles became rounded and ligatures developed. These tendencies were completed during the Persian period, and they are emphasized in the Aramaic writing on papyrus employed in Egypt between 50o and 200 B.C. Other developments of Ara Greek, and an Armenian Christian literature thus arose. It is likely that the origin is Greek, a few Persian characteristics that appear being due to the influence of the dominating power.
In all Teutonic countries are found inscriptions in the alphabet known as Runic. The affinities of this alphabet with the Greek are apparent, but there has been dispute as to whether it is de rived directly from Greek or through an Italic alphabet. The strongest argument in favour of a Latin derivation is the fact that the value of the letter ri, is f, not w. L. F. A. Wimmer in his book Die Runenschri f t (1887) contended strongly for a Latin origin. He believed that the Runes developed from the Latin alphabet at the end of the end century A.D. This can scarcely be the case however, for the Romans had been in contact with the Germanic tribes for some time before this date, and the alpha bet, if borrowed from them, would thus have been borrowed sooner. Again many Runic inscriptions are written from right to left or jovorpok8bv and this would have been impossible had the alphabet been borrowed from one uniformly written from left to right. Others have concluded with more probability that the Runes are derived directly from a western Greek alphabet resembling that which appears on the Formello abcdaria. It is possible that knowledge of such an alphabet reached northern Europe from Italy through the medium of Etruscan traders. This might account for the value of Isaac Taylor's view was that the alphabet was derived from a Greek colony on the Black Sea about 600 B.C. Of the two views this is perhaps the more probable. It may even be that the alphabet was derived in the region of the Black Sea not directly from a Greek, but from an Asianic source, closely parallel with the Greek developments, as we have already seen the Etruscan alphabet probably to have had its origin. There is some, but not conclusive, evidence that the alphabet was borrowed previous to the Teutonic sound-shift.
An outcome of the Runic alphabet was the cryptographic Ogham writing employed by Celtic peoples in Britain and Ireland (see OGHAM LANGUAGE). In this the several letters of the alpha bet were represented by strokes differing in number (from one to five) and in position on either side, or on both sides, of a straight horizontal line.
maic are modern square Hebrew and the alphabet of Palmyra.
A group of alphabets arose in Asia Minor during the first half of the first millennium B.C. deriving from a common source, probably an early stage of Aramaic. The Greek alphabet in its two varieties was one of these. The Carian alphabet was another, as also the Lycian. These both seem parallel to Greek, not off shoots of it. The Phrygian on the other hand seems to be derived from the Greek, and from the western variety.
The Armenian alphabet owes its origin to Bishop Mesrob late in the 4th century A.D. Both an Iranian and a Greek origin have been claimed for it, Isaac Taylor contending for the former, Gardthausen and Hubschmann for the latter. The country had been divided between Persia and the Byzantine empire. The larger share fell to the Persians who discouraged the study of The date of these inscriptions in Wales is the 5th and 6th centuries A.D.
Finally must be mentioned the two Slavonic alphabets, Cyrillic and Glagolitic, both according to tradition the invention of Cyril, the missionary to the Slays, in the 9th century A.D. Each is taken from the Greek alphabet of that period, Glagolitic from the cursive writing, Cyrillic from uncial, when the value of B had become the spirant v, that of H i, of c13 f and of X the sound of Scotch ch. Several additional symbols were invented to express those sounds of the Slavonic language which could not be com prised within the compass of the Greek alphabet. Glagolitic survived in Croatia till the i 7th century. Cyrillic has for long been the alphabet in which the Slavonic languages are normally expressed.
AUTHORITIES.--General: Faulmann, Illustrierte Geschichte der Authorities.--General: Faulmann, Illustrierte Geschichte der Schrift (188o) ; Isaac Taylor, The Alphabet (1883) (2nd ed., 1899) 2 vols.; Philippe Berger, Histoire de l'ecriture dans l'antiquite (1891) (2nd ed., 1892) ; E. Clodd, The Story of the Alphabet.
Sinai Inscriptions: Alan H. Gardiner, "The Egyptian Origin of the Semitic Alphabet," in Journal of Egyptian Archaeology, 3 (1916), p. 1 ff.; Hans Bauer, Zur Entzi ff erung der neuentdeckten Sinaischri f t, Halle (1918) .
South Semitic: D. H. Muller, Epigraphische Denkmaler aus Arabien (1889) ; Hommel, Siid-arabische Chrestomathie (1893) ; D. H. Muller, Epigraphische Denkmaler aus Abessinien (1894) ; Fr. Pratorius, Zeit schrif t der deutschen morgenlandischen Gesellschaft, vol. 58 (19o4) pp. Origin of Phoenician Alphabet; De Rouge, Memoire sur l'origine egyptienne de l'alphabet phenicien (1874) ; Gesenius-Kautzsch, Hebraische Grammatik, 28th ed. 5g (pp. ; J. P. Peters, "Recent Theories of the Origin of the Alphabet" in Journal of the American Oriental Society, vol. 22 (19o1) (pp. 177-198) ; Noldeke, Beitrage zur Semitischen Sprachwissenschaft (Strassburg, 1904) (pp. 124-136) ; Fr. Pratorius, Ober den Ursprung des kanaanaischen Alphabets (1906). Translation in Annual Report of the Board of Regents of the Smithsonian Institution for (pp. R. Dussaud, Les Arabes en Syrie avant l'lslam (1907) (pp. A. H. Sayce, "The Origin of the Phoenician Alphabet" in Proceedings of the society of Biblical Archaeology, vol. xxxii. (191o) (pp. 215— 2 215 222) ; Sir W. Flinders-Petrie, The Formation of the Alphabet (1912) ; S. Macalister, The Philistines (1914) (pp. .
North Semitic: Lidzbarski, Nordsemitische Epigraphik (1898), Jewish Encyclopedia, vol. i. (1901), and table of symbols in 27th edition of Gesenius' Hebraischer Grammatik (1902) .
Greek: Salomon Reinach, Traite d'epigraphie grecque (1885) ; A. Kirchhoff, Studien zur Geschichte des griechischen Alphabets (4th ed. 1887) ; E. S. Roberts, Introduction to Greek Epigraphy (pt. i., 1887), with E. A. Garr finer (pt. ii., 1905) ; W. Larfeld, "Griechische Epigraphik" in Handbuch der klassischen Altertumwissenschaft (1914) .
Lydian: Pauly-Wissowa, Real-Encyklopadie vol. xiii. (1927), article Lydia, p. 2,157 f.
Latin: Emil Hubner in Miiller's Handbuch der klassischen Alter tumwissenschaft.
Etruscan: Pericle Ducati, Etruria Antica, vol. i., p. 67 f. (1925) ; D. Randall Maclver, "The Etruscans" in Antiquity, vol. i., No. ii. (June 1927) p. 159 ff.
Indian: Cambridge History of India, vol. i. (edited by Prof. E. J. Rapson) p. 62 ; G. Baler, Indische Palaeographie, Grund d. indo-ar. Phil. (Strassburg, 1896) : English translation, Indian Palaeography, ed. by J. F. Fleet, Ind. Ant., 1904 (App.). Pahlavi: Ernst Herzfeld, Paikuli (1924) . Runes: Isaac Taylor, Greeks and Goths; R. M. Meyer in Paul Braune and Siever's Beitrage, xxi. (1896) pp. 162 ff.; Hempl in papers in Philologische Studien (1896) and in Journal of Germanic Philology, ii. pp. 37o ff.; Gundermann in Literaturblatt fur germanische and romanische Philologie (1897) col. 429 f.; L. F. A. Wimmer, Die Runenschrift (1897). Oghams: Rhys, Outlines of Manx Phonology, p. 73 (Publications of the Manx Society, vol. xxxiii.) ; Rhys and Brynmor Jones, The Welsh People pp. 3, 502 ; Whitley Stokes in Bezzenberger's Beitrage, xi. (1886) p. 143 ff.; R. A. S. Macalister in Studies in Irish Epigraphy, 1897, 1902, 1907. Cyrillic and Glagolitic: Leskien, Grammatik der altbulgarischen (altkirchen slavischen) Sprache (Heidelberg, 1909) ; further references will be found from the introduction. Isaac Taylor, Archiv f iir slavische Philologie, v., 191 ff. For excellent comparative tables of most known Alphabets see F. Ballhorn, Alphabete orientalische and occidentalische Sprachen (Leipzig, 1859). (B. F. C. A.)