[This entry treats the origins and development of what is traditionally termed the “alphabet,” that is, the stages termed “abjad” and “alphabet.” For definitions and further discussion, see Writing and Writing Systems.]

The alphabet began as a system of symbols, each of which represented a consonantal phoneme. Because in the earliest attested stages the signs are quasi-pictographic and the form of the sign corresponds in several cases to the name of the letter known from later periods, it is plausible that each of the earliest sign forms bore a name beginning with the phoneme represented by the sign. For example, the name of the {b}-sign would have been baytu, the Semitic word for “house,” and it would have represented the consonantal phoneme /b/. This is the so-called acrophonic principle.

More than a millennium after writing had come into use in Mesopotamia and in Egypt, the alphabetic principle was devised somewhere in the Levant between these two early centers of civilization. Because of the representational character of the earliest signs, the invention probably took place under Egyptian rather than under Mesopotamian influence, for hieroglyphs maintained pictographic forms for millennia, even while cursive scripts developed. [See Egyptian; Hieroglyphs.] On the other hand, although a system of quasi-alphabetic signs was developed in Egyptian, used principally to represent foreign names, each Egyptian sign continued to note one, two, or three consonants until the Greek alphabet was adopted for Coptic. [See Hieroglyphs; Greek; Coptic.] The earliest alphabet clearly did not originate in a simple borrowing of the Egyptian quasi-alphabetic signs, for no single correspondence between the two systems has been proven. Rather, even though the idea may well have been born under the influence of the Egyptian usage (Sznycer, 1974; Sass, 1991), the true invention is the idea of noting the consonantal segment only, for that resulted in the radical reduction of the number of signs: from the large number required for a syllabic system to a much small number required for an alphabetic system. It is estimated that a working Mesopotamian scribe needed an active knowledge of approximately two hundred signs, but the earliest alphabets probably did not have more than twenty-nine signs at the most. The imitation of the Egyptian iconic principle of writing also resulted in a set of relatively simple symbols; the Mesopotamian cuneiform system, by contrast, included many signs that consisted of a large number of wedges (ten or more in the archaic period) and that required much practice to produce and to recognize. [See Cuneiform.] It can be debated whether the consonantal writing system merits the name alphabet, but the simplification of the writing process it represented makes it the decisive step toward the full alphabet as devised by the Greeks. [See Writing and Writing Systems.]

Because of the small number of extant early alphabetic inscriptions and the difficulties in dating them, it is uncertain just where the alphabet was invented. Because the earliest known examples are arguably the Proto-Sinaitic texts, it is tempting to see in them the texts for the writing of which the alphabet was invented. That temptation must be resisted, however, for it is presently impossible to determine whether the social conditions requisite for the invention of a writing system existed in the Sinai then—that is, the presence of a scribe of the language for which the alphabet was invented who was also well acquainted with Egyptian. It is in any case likely that the invention was only accomplished once and spread throughout the Levant. In theory, it could have occurred anywhere there was an Egyptian influence in the late Middle Kingdom—from Egypt itself through Canaan into Syria. The concentration of early texts in Sinai and the various inscriptions from sites in southern Canaan, however, make the most plausible hypothesis that of a southern origin.

If the hypothesis be accepted that the above Proto-Sinaitic and Proto-Canaanite inscriptions represent the earliest alphabetic writings, a necessary corollary is that the alphabet was invented to write a West Semitic language. [See Proto-Sinaitic; Proto-Canaanite.] Unfortunately, problems of overall decipherment and of the identification of individual signs and their variants make it impossible at present to determine the precise characteristics of the language(s) represented by the earliest inscriptions. The data available indicate an early dialect of Canaanite.

Ugaritic Alphabet.

The earliest West Semitic language represented in an alphabetic script and with an important body of decipherable inscriptions is Ugaritic. [See Ugarit; Ugaritic; Ugarit Inscriptions.] These texts date to the fourteenth-thirteenth centuries BCE, although the greatest number are probably from the last few decades before the destruction of the site in the early twelfth century BCE.

Ugaritic made two major contributions to the early history of the alphabet: it is the only West Semitic alphabetic writing system that is cuneiform and a significant number of abecedaries exist (an abecedary is the conventional name for the writing out of an alphabet according to an established order).

For reasons presently unknown (the prestige of Mesopotamian cuneiform, a temporary shortage of papyrus?), the West Semitic alphabet was reduced to wedge forms and written on clay at Ugarit. [See Papyrus; Writing Materials.] Various considerations—chronological, formal, and structural—make it highly unlikely that the Ugaritic writing system represents a local invention of the alphabet: it appeared too late (fourteenth century BCE), there are several plausible correspondences of sign forms between it and the known linear writing systems, and it does not correspond perfectly to the Ugaritic phonological system.

The Ugaritic abecedaries have established several valuable reference points in the history of the alphabet.

  • • They prove that the order of the letters later known as the standard Northwest Semitic order (Hebrew, Phoenician-Punic, Aramaic) was in use in the fourteenth century BCE in Syria. [See Hebrew Language and Literature; Phoenician-Punic; Aramaic Language and Literature.] Abecedaries are attested in Palestine and in Phoenicia by the late thirteenth century BCE and on into the first millennium BCE (Cross, 1980, pp. 8–15; Lemaire, 1978). The early date of the Ugaritic examples and the wide spread of the later languages attesting this order make it plausible that the inventor of the alphabet had conventionally used this order for pedagogic purposes.
  • • Comparison between the Ugaritic abecedaries and the order of letters in the later Northwest Semitic languages shows that a “long” alphabet was in use in the fourteenth century BCE. The basic consonantal inventory of Ugaritic was represented by twenty-seven signs, whereas the basic southern Canaanite alphabet consisted of twenty-two signs. The five extra Ugaritic signs are interspersed among the signs known from the southern Canaanite alphabet as follows:

Canaanite b g d h w z y k
Ugaritic b g d h w z y k š
Canaanite l m n s p q r š t
Ugaritic l m n s p q r ǵ t (ỉ ủ )
  • • If the inventor of the Ugaritic cuneiform system had been imitating a “short” alphabet, he would in all likelihood have tacked the extra signs he needed on the end of the alphabet, as he did with the three extra signs that he did devise ({ỉ, ủ, s̀})—and as the Greeks did when they borrowed and adapted the West Semitic alphabet; or he would have adapted existing signs in groups, as the Arabs did when they revised the Northwest Semitic alphabet. Rather, he seems to have been following an established order in which {ḫ} followed {g}, {š} followed {k}, et cetera. Because the Ugaritic phonological system does not perfectly match the writing system (e.g., is used inconsistently), it is likely that the alphabet was originally borrowed from a West Semitic language with a slightly different consonantal inventory and/or that the use of the long linear alphabet already had a history at Ugarit.
  • • The order of the long alphabet, as illustrated by the Ugaritic abecedaries, makes the hypothesis plausible that the short alphabet is a simplification of the longer one: that is, certain sign forms were dropped from usage when not needed (probably in the process of adoption from one language to another, rather than by alphabet reform in a given community), for alphabet usage tends to be conservative.
  • • The recent discovery at Ugarit of an abecedary in the South Semitic order and with variant sign forms (Caquot, forthcoming; Bordreuil and Pardee, forthcoming), and the earlier decipherment of a cuneiform tablet from Beth-Shemesh as an abecedary in the South Semitic order (Loundin, 1987; further bibliography in Bordreuil and Pardee, forthcoming), prove the widespread use of a different alphabetic order coupled with variant sign forms. [See Beth-Shemesh.] These abecedaries must reflect the use of a language of the South Semitic type in Canaan, although they and a few signs on sherds from Kamid el-Loz (Röllig and Mansfeld, 1969–1970) are the only present evidence for that usage. [See Kamid el-Loz.]

Spread of the Semitic Alphabet.

The date of the earliest Phoenician inscriptions has been much debated. The Ahiram inscription is the earliest continuous text in a Phoenician dialect, but arrowhead inscriptions from about 1100–900 BCE and a few very brief inscriptions from Byblos and other sites illustrate the use of the alphabet in the Early Iron period. [See Ahiram Inscription; Byblos.] It is also clear that the Ugaritic cuneiform system was occasionally used to write Canaanite at the very end of the Late Bronze Age (the language of a late thirteenth-century BCE inscription from Sarepta in Ugaritic script has been classified as Phoenician: see Greenstein, 1976, and Bordreuil, 1979). [See Sarepta.]

By the late tenth–early ninth centuries BCE, inscriptions in Aramaic and Hebrew illustrate the spread of the Phoenician alphabet inland. It is clear that the Aramaeans borrowed the alphabet from the Phoenicians because the sign forms in the earliest inscriptions are Phoenician and because there were major phonological differences between the two languages that affected the development of Aramaic orthography. [See Arameans; Phoenicians.] For example, /ḍ/ and /ṣ/ had coalesced in Proto-Phoenician and only one sign remained; Aramaic, on the other hand, had a phoneme, /ḍ/, the scribes represented by means of {q}; at a later point, the pronunciation of /ḍ/ shifted and the scribes began representing it by {῾}: for example, “earth” {'rṣ} in Phoenician is written {'rq} in Old Aramaic, {'r῾} in the Persian period.

Because /ś/ survived as a phoneme in Hebrew but without a corresponding sign in the Hebrew alphabet, it is clear that the Hebrews borrowed the standard short alphabet from the Phoenicians or from speakers of another Northwest Semitic language who used the same alphabet. (For the possibility that other consonantal phonemes had survived in the Hebrew of the biblical period that were not noted in the writing system, see Kutscher, 1982, pp. 17–18.) All that can be said for the present is that the Hebrew alphabet does not represent an independent adaptation of the short alphabet to the peculiarities of the Hebrew language (cf. Cross, 1980, pp. 13–15).

Within the Levant the various linguistic/political/geographic entities adopted the alphabet, and the sign forms underwent local evolutions (Naveh, 1982/1987; Cross, 1980). The alphabet spread throughout the Semitic world and to its colonies (Phoenicians in the West). After Aramaic was adopted as a lingua franca by the Achaemenids, its usage spread throughout the Near East and beyond; Aramaic even seems to have been used logographically in areas corresponding to modern Turkmenistan and Georgia (Naveh, 1982/1987, pp. 127–130). The short alphabet was eventually borrowed by Arabic speakers and adapted in a different order to express the fuller consonantal inventory of Arabic—twenty-eight phonemes (Naveh, 1982/1987, pp. 153–162). [See Arabic.]

Contribution of Greek.

There has been a long controversy about the date at which the Greeks borrowed the West Semitic consonantal alphabet, with proposals ranging over several centuries—from the middle of the second millennium BCE to the early eighth century BCE. Phoenician and Aramaic inscriptions from as early as about 900 BCE (Bisi, 1991) are rare in the West, and the earliest Greek inscriptions date to the early eighth century BCE; thus, a hypothetical earlier borrowing of the alphabet must be based on the typology of letter forms: the earliest Greek letters most clearly resemble West Semitic signs from what period? Viewed abstractly, denying a date significantly earlier than the first inscriptions is essentially an argument from silence: because no examples exist of Greek writing earlier than the eighth century BCE, the writing system was adopted relatively shortly before that date. Following Rhys Carpenter's 1933 article of fundamental importance, P. Kyle McCarter (1975) and Maria Giulia Amadasi Guzzo (1991) have made very strong arguments from sign typology that essentially back up the argument from silence, to the effect that archaic Greek letters look like West Semitic forms from the late ninth/early eighth centuries BCE.

A less heated debate has been carried on with regard to the place where the alphabet was borrowed, with proposals tending to fit the date being argued. Because of the varieties of letter forms and alphabetic configurations attested among the Greeks, there are even those who favor multiple borrowings: while refusing to choose a particular place of borrowing, Amadasi Guzzo (1991, p. 309) makes an essential methodological point: the sophistication shown in recognizing phonetic equivalences between the Greek borrower's language, the Phoenician system being borrowed, and the adaptations to Greek needs requires that the place of borrowing be a metropolitan center with significant groups of both Greeks and Phoenicians, including well-trained scribes. [See Scribes and Scribal Techniques.]

The alphabet borrowed by the Greeks was of the Phoenician type, consisting of twenty-two consonantal signs. This is clear both from early Greek abecedaries and from the numerical value of the Greek letters (Piérart, 1991, pp. 571–573, figs. 1, 2). Most scholars assume that the Phoenicians themselves were the transmitters—in no small part because of the many contacts between Greeks and Phoenicians on their common Mediterranean waters. It cannot be ruled out that the Arameans were involved, but because the Arameans were not known as seafarers, this theory usually involves overland transmission across Asia Minor (Amadasi Guzzo, 1991). The strongest basis for this theory remains the Greeks names of the letters, such as alpha and delta, for they seem to reflect Aramaic pronunciation, not Phoenician (which would be 'alp and dalt). It has been counterargued that forms like alpha only represent the Greek pronunciation of consonantal clusters such as /-lp/ at the end of the word, whereas some letter names, such as iota, presuppose a non-Aramaic origin (Naveh, 1982/1987, p. 183).

By inventing the notation of vowels, the Greeks made their own contribution to the history of the alphabet. On the Semitic side, a start in this direction had been made that has become systematized in Hebrew and in Arabic: the use of matres lectionis. The consonants 'aleph, hē, wāw, and yōd were first used by Arameans and Hebrews to indicate a long vowel. The usage may have arisen by analogy from historical writings (e.g., if/bayt/becomes/bêt/, the tendency is to continue writing the word with {y}, even though the consonant is no longer pronounced). It never became a rigid system in any of the Hebrew or Aramaic dialects of the pre-Christian era, but in time more and more matres lectionis are found in the texts. They are used, for example, more frequently in the so-called sectarian documents from Qumran than in the received text of the Pentateuch, which represents a stage of the text dating a few centuries earlier than the Dead Sea Scrolls. [See Qumran; Dead Sea Scrolls.]

It is for the present a moot point whether the Greeks knew about such usages—they certainly did not if the alphabet was borrowed from the Phoenicians, for matres lectionis were not a part of normative Phoenician orthography in the eighth century BCE. What the Greeks did, in any case, was to give vocalic values to certain Phoenician consonants that did not correspond to consonantal phonemes in the Greek language, beginning with the first letter of the alphabet: the sign for the West Semitic consonant /'/ was used to represent the vowel /a/ by the Greeks. The signs for all the Semitic gutturals (/', h, ḥ, ῾/) were so adapted, as were those for the two continuants /w/ and /y/.

The addition of vowel signs made the alphabet more explicitly expressive, for it provided a more complete pronunciation of a word than did the notation of consonants only. The notation of vowels made the Greek alphabet a stronger pedagogic tool, allowing for a decrease in the importance of oral tradition in teaching. This increased expressivity not only gave a more complete rendition of words in a given language, but also facilitated reading and understanding a foreign language. Greek also had a series of consonantal phonemes that did not have correspondences in Phoenician. New signs were invented or existing ones were transmuted for those sounds and they were added to the end of the alphabet—whence come several of the letters that occur in the English alphabet after {t}, the last sign of the West Semitic alphabet. The transmission of the alphabet to the Greeks permitted the writing down of Greek traditions that have by this means become part of modern culture. For example, the Homeric traditions plausibly began being committed to writing in the eighth century BCE, shortly after the full alphabet was devised (Amadasi Guzzo, 1991, pp. 307–308; Powell, 1991).

The invention of the consonantal alphabet by the Semites and its transformation by the Greeks for the notation of vowels are often cited as important factors in the democratization of writing. Although widespread literacy did not follow immediately upon the invention of the alphabet, a more rapid spread of literacy can be seen after the invention of the vowels. [See Literacy.] There is certainly a good deal of truth in linking democratization and the alphabet: to the extent that the growth of democracy required an educated middle class to deal with the complexities of growing societies, the ease of use of the alphabet may be thought to have facilitated, by the principle of efficient use of resources, this process.

Owing to the spread of alphabetic usage throughout the Mediterranean world and eastward during the last centuries of the pre-Christian era, the major writings of the three great Western religions have, with relatively few exceptions (e.g., Ethiopic), been transmitted in alphabetic scripts (e.g., Hebrew, Greek, Latin, Syriac, Coptic, Arabic, Armenian).

[See Ethiopic; Latin; Syriac; Coptic; Armenian.]


  • Amadasi Guzzo, Maria Giulia. “‘The Shadow Line’: Réflexions sur l'introduction de l'alphabet en Grèce.” In Phoinikeia Grammata, lire et écrire en Méditerranée: Actes du colloque de Liège, 15–18 novembre 1989, edited by Claude Baurain et al., pp. 293–311. Namur, 1991. Excellent, brief overview of the origin of the alphabet and its transmission to the Greeks, with a good presentation of methodological factors. The essential bibliography is provided in footnotes.
  • Bisi, Anna M. “Les plus anciens objets inscrits en phénicien et en araméen retrouvés en Grèce: Leur typologie et leur rôle.” In Phoinikeia Grammata, lire et écrire en Méditerranée: Actes du colloque de Liège, 15–18 novembre 1989, edited by Claude Baurain et al., pp. 277–282. Namur, 1991. Discusses the implications of the presence of Phoenician- and Aramaic-transcribed objects in the West.
  • Bordreuil, Pierre. “L'inscription phénicienne de Sarafand en cunéiformes alphabétiques.” Ugarit Forschungen 11 (1979): 63–68. Improved reading of the Sarepta inscription in Ugaritic script.
  • Bordreuil, Pierre, and Dennis Pardee. “Les textes ougaritiques.” Forthcoming in the edition of texts discovered at Ras Shamra from 1986 to 1992. Includes the editio princeps of the abecedary in South Semitic order.
  • Caquot, André. “Un abécédaire du type sud-sémitique découvert en 1992 dans les fouilles archéologiques françaises de Ras Shamra-Ougarit.” Comptes Rendus de l'Académie des Inscriptions et des Belles Lettres. Forthcoming. First announcement of the discovery of an abecedary of South Semitic type at Ras Shamra.
  • Carpenter, Rhys. “The Antiquity of the Greek Alphabet.” American Journal of Archaeology 37 (1933): 8–29. Fundamental article on the methodology of dating the borrowing of the alphabet by the Greeks. More recent discoveries have moved the date a bit earlier, but the basic arguments still stand.
  • Cross, Frank Moore. “Newly Found Inscriptions in Old Canaanite and Early Phoenician Scripts.” Bulletin of the American Schools of Oriental Research, no. 238 (1980): 1–20. Reappraisal of alphabetic origins and developments which accepts an early borrowing of the alphabet by the Greeks (eleventh century).
  • Greenstein, Edward L. “A Phoenician Inscription in Ugaritic Script?” Journal of the Ancient Near Eastern Society of Columbia University 8 (1976): 49–57. Identifies the language of the Sarepta inscription in Ugaritic script as Phoenician.
  • Kutscher, Eduard Y. A History of the Hebrew Language. Jerusalem, 1982. Includes a discussion of the relationship of graphemes to phonemes in Biblical Hebrew.
  • Lemaire, André. “Fragment d'un alphabet ouest-sémitique du VIIIe siècle av. J.-C.” Semitica 28 (1978): 7–10. Edition of a partial abecedary purchased on the antiquities market.
  • Lunden [Loundine], A. G. “L'abécédaire de Beth Shemesh.” Le Muséon 100 (1987): 243–250.
  • McCarter, P. Kyle, Jr. The Antiquity of the Greek Alphabet and the Early Phoenician Scripts. Harvard Semitic Monographs, 9. Missoula, 1975. The basic work accepting a late borrowing of the alphabet by the Greeks (though it leaves open the possibility of earlier experiments). Competent treatment of both the Semitic and the Greek sides of the question.
  • Naveh, Joseph. An Early History of the Alphabet: An Introduction to West Semitic Epigraphy and Palaeography. 2d ed. Jerusalem, 1987. Good overview of alphabetic origins and developments, though one may query the late dating of the Proto-Sinaitic inscriptions (fifteenth century) and the early borrowing of the alphabet by the Greeks (eleventh century). Generally more thorough on graphic evolutions than on correlations between graphemes and phonemes.
  • Piérart, Marcel. “Écriture et identité culturelle: Les cités du Péloponnèse nord-oriental.” In Phoinikeia Grammata, lire et écrire en Méditerranée: Actes du colloque de Liège, 15–18 novembre 1989, edited by Claude Baurain et al., pp. 565–576. Namur, 1991. Discussion of the contribution of Greek abecedaries to the early history of the alphabet among the Greeks.
  • Powell, B. P. “The Origins of Alphabetic Literacy among the Greeks.” In Phoinikeia Grammata, lire et écrire en Méditerranée: Actes du colloque de Liège, 15–19 novembre 1989, edited by Claude Baurain et al., pp. 357–370. Namur, 1991. Sees the origin of vowel notation by the Greeks in the requirements of representing the hexameter in writing.
  • Röllig, Wolfgang, and G. Mansfeld. “Zwei Ostraka vom Tell Kamidel-Loz und ein neuer Aspekt für die Entstehung des kanaanäischen Alphabets.” Die Welt des Orients 5 (1969–1970): 265–270.
  • Sass, Benjamin. Studia Alphabetica: On the Origin and Early History of the Northwest Semitic, South Semitic, and Greek Alphabets. Orbis Biblicus et Orientalis, 102. Freiburg, 1991. Overview of various crucial points in the origin and transmission of the alphabet.
  • Sznycer, Maurice. “Quelques remarques à propos de la formation de l'alphabet phénicien.” Semitica 24 (1974): 5–12. Brief but systematic treatment of alphabetic origins.

Dennis Pardee