Linear B (Mycenaean Greek)

Overview

Linear B is a syllabic writing system of the Aegean Bronze Age (ca. 1450–1200 BC) used to record Mycenaean Greek, the earliest known form of the Greek language, predating Homer by roughly seven centuries. It is a strictly administrative and accounting script, employed by the scribes of the Mycenaean palaces (Knossos in Crete; Pylos, Mycenae, Tiryns, and Thebes on the Greek mainland) to keep the economic records of the palatial power: inventories of goods, rations, personnel, flocks, offerings, and contributions. No literary, narrative-religious, or epistolary texts have survived — the corpus is entirely bureaucratic. The script derives from Minoan Linear A (undeciphered) and was adapted to transcribe a different, Indo-European language. Its decipherment by Michael Ventris and John Chadwick in 1952–1953 stands as one of the great intellectual achievements of the twentieth century; it proved the continuity of Greek civilization since the Bronze Age, radically transforming our understanding of Greek prehistory and of the Homeric epic.

History of the people

The Mycenaeans are the Greeks of the Late Bronze Age (Late Helladic / Mycenaean civilization, ca. 1600–1100 BC), named after the site of Mycenae excavated by Heinrich Schliemann. Organized into independent palatial kingdoms centered on fortified citadels (Mycenae, Tiryns, Pylos, Thebes, Orchomenos, Athens) and, after their seizure of Knossos (supplanting the Minoan administration), on Crete as well, they formed a highly hierarchical and bureaucratic society. At the summit were the wa-na-ka (wanax, king) and the ra-wa-ke-ta (lawagetas, "leader of the people/army"); below them, an elite of te-re-ta (telestai, landholding notables) and e-qe-ta (hequetai, "companions/followers"), an administration of scribes, local collectivities (da-mo, dāmos), and dependents including the do-e-ro/do-e-ra (dowelos/dowela, "slaves," often "slaves of a god"). The palatial economy controlled agriculture, animal husbandry (large sheep flocks at Knossos for wool), luxury craft production (ivory, bronze, textiles, perfumes), and Aegean trade. The tablets also document a pantheon that includes already-recognizable Greek deities: di-we (Zeus), po-se-da-o (Poseidon), e-ra (Hera), a-ta-na (associated with Athena), di-wo-nu-so (Dionysus). Around 1200 BC, widespread destructions, the collapse of the palaces, and the end of the Mycenaean world ensued.

History of the language

Mycenaean Greek represents the earliest attested phase of Greek, spoken during the Mycenaean floruit (ca. fifteenth–thirteenth centuries BC). The tablets come from the palaces' destruction levels: Knossos (debated chronology, often placed in the fourteenth century BC, though some see later deposits there), Pylos, and mainland sites around ca. 1200 BC (end of Late Helladic IIIB). Writing ceases entirely with the collapse of the palaces around 1200–1180 BC: purely bureaucratic, Linear B vanishes together with the palatial apparatus that sustained it. There follow the "Dark Ages" (ca. 1100–800 BC), a period with no attested writing, until the adoption of the alphabet derived from Phoenician (eighth century BC). The Greek language does NOT die out: it survives orally and evolves into the dialects of the first millennium. Mycenaean is not the direct ancestor of any single Classical dialect; it shares the most traits with the Arcado-Cypriot group. The Cypriot syllabary of the first millennium belongs to the same Aegean writing tradition but is not directly descended from Linear B.

The script

Linear B comprises three categories of signs. (1) A SYLLABARY of about 87 phonetic signs recording open syllables: pure vowels (a, e, i, o, u) and consonant+vowel syllables (the d, j, k, m, n, p, q [labiovelars], r, s, t, w, z series), plus complex/double signs (a2 = ha, a3 = ai, ra2, ta2, pa3…) and signs of uncertain value (the so-called "optional signs"). (2) LOGOGRAMS or ideograms (about a hundred), drawings that directly represent the goods being counted: categories of persons (VIR = man, MUL = woman), animals (BOS, OVIS, SUS, CAP, EQU), commodities (GRA = wheat/grain, HORD = barley, VINum, OLEum = oil, FICus = figs), metals (AURum, AES = bronze), textiles, vessels, chariots, wheels, weapons. (3) A decimal NUMERICAL system (units, tens, hundreds, thousands, tens of thousands) and distinct METROLOGICAL signs for measures of capacity (dry and liquid) and of weight. The script reads from left to right, in horizontal lines drawn with ruling; words are separated by a short vertical stroke (word-divider). The principal medium is the unbaked clay tablet, dried and then inscribed with a stylus, preserved by accidental firing during the destruction fires of the palaces. Painted inscriptions on vases ("stirrup jars") are also found. The tablets have standardized formats (the elongated "palm-leaf," or the rectangular "page").

Sign system

The basic syllabary is organized in a table (a "grid") cross-referencing consonants and vowels, a structure that Ventris reconstructed even before knowing the phonetic values. Pure vowels: a, e, i, o, u. Examples of secure and frequent values: da/de/di/do/du, ka/ke/ki/ko/ku, ma/me/mi/mo/mu, na/ne/ni/no/nu, pa/pe/pi/po/pu, ra/re/ri/ro/ru, sa/se/si/so/su, ta/te/ti/to/tu, wa/we/wi/wo, qa/qe/qi/qo (the labiovelar series, denoting kʷ/gʷ/kʷʰ), za/ze/zo, ja/je/jo. Notable signs heavily used in the corpus: ko-no-so (Knossos), pu-ro (Pylos), wa-na-ka (wanax, king), qe ("and," enclitic = Greek τε). The standard numbering of the signs follows the system established by E. L. Bennett Jr. and adopted by Ventris (transnumeration with the prefix *, e.g. *08 = a, *61 = o; da = *01, etc.), which remains the reference for designating signs, particularly those whose phonetic value is uncertain or debated. The logograms are designated by conventional Latin abbreviations (VIR, MUL, GRA, VIN, OLE, etc.). Some syllabograms also serve as an "acrophonic" base or as monograms/ligatures modifying the logograms (e.g. categories of wool and cloth).

Phonology

The language recorded is Mycenaean Greek, an archaic dialect preserving features later lost: labiovelars (kʷ, gʷ, kʷʰ; the q series) and /w/ (digamma; the w series, e.g. wa-na-ka). Vowels a/e/i/o/u (length not marked); voiceless/voiced/aspirated stops (p/b/pʰ, t/d/tʰ, k/g/kʰ). But the syllabary records this phonology very imperfectly: it does not distinguish voiceless/voiced/aspirated (EXCEPT the dental series, which opposes d to t), nor l and r (a single r series), and it cannot directly write closed syllables or consonant clusters. Orthographic consequences: (1) a syllable-/word-final consonant is generally omitted (ka-ko = khalkos "bronze"; pa-te = patēr); (2) within an internal cluster, each consonant receives a "dead" vowel copied from a neighboring vowel (a-to-ro-qo = anthrōkʷos; ti-ri-po = tripos "tripod"); (3) /h/ is not marked except a2 = ha; (4) nasals before a consonant are often omitted; (5) /s/ before a consonant and final -s are often omitted. Hence a strong ambiguity and multivalence in the readings.

Grammar — essentials

Mycenaean Greek is an Indo-European inflectional language (an archaic state of Greek). Salient features, despite the accounting nature and fragmentary state of the texts: preservation of a form in -pi (from the Indo-European instrumental/comitative *-bhi, e.g. e-re-pa-te-jo-pi "of ivory," functioning chiefly in the plural and combining instrumental/locative values) and of a LOCATIVE in -i, alongside the nominative, genitive, dative, and accusative. Thematic GENITIVE singular in -o-jo (= -osyo, older than Classical -ou). DATIVE plural in -si; forms in -pi for certain stems. Retention of the LABIOVELARS (the q series) where first-millennium Greek transforms them (qe = "and"). Numerous ADJECTIVES of material/belonging in -e-jo/-e-ja. The verb is sparsely attested: e-ke (ekhei "he has"), e-ko-si ("they have"), o-pe-ro ("owed, lacking"), di-do-si ("they give"), participles in -me-no. Augment rare/absent. The "syntax" is essentially a list syntax: entry (person/place/good) + logogram + numeral. Toponyms and anthroponyms dominate the lexicon.

Reading tips

Operational reading method. (1) Segment the text according to the word-dividers and identify the "entry + logogram + numeral" structure typical of accounting tablets. (2) Transliterate each syllabogram by its conventional value in lowercase joined by hyphens (e.g. ko-wo, a-to-ro-qo), using the *NN numbering for any sign of doubtful value. (3) Reconstruct the Greek word by applying the defective orthographic rules IN REVERSE: restore the omitted syllable-final consonants, merge the echo-vowels of consonant clusters, restore unwritten /h/, /s/, and nasals, test l/r (the ambiguous r series) and voiceless/voiced/aspirated (except dentals). (4) Cross-check against the attested Mycenaean lexicon and the neighboring ideograms: the logogram often disambiguates (VIR/MUL indicates a person, OVIS a sheep). (5) Distinguish toponyms, anthroponyms (very numerous, often untranslatable), and vocabulary words. (6) Pay attention to function words: qe "and" (final enclitic), -de, o- at the head of a rubric, pa-ro "from, in the possession of," o-pe-ro "deficit/owed," to-so/to-sa "in total so many." (7) Always propose the sign-by-sign transliteration first, THEN the interpretation, explicitly signaling the possible multiple readings and the uncertain restorations. Never force a "literary" translation: these texts are registers.

Pitfalls

Major pitfalls of reading and interpretation. (1) DEFECTIVE orthography: syllable-final consonants omitted, consonant clusters broken up by echo-vowels, /h/, /s/, nasals, and final -s/-n often unwritten → a single spelling may correspond to several Greek words. (2) NON-DISTINCTION of voiceless/voiced/aspirated (except dentals d/t) and l/r → structural phonological ambiguity. (3) Vowel length never marked. (4) Massive presence of ANTHROPONYMS: many sequences are proper names that cannot be translated; do not force a lexical meaning at all costs. (5) Numerous LOGOGRAMS and metrological abbreviations whose exact value remains disputed. (6) Signs of UNCERTAIN value (the *NN series): do not invent a value. (7) LACUNARY texts (broken tablets): editorial brackets/dots signal hypothetical restorations that must not be treated as certain. (8) Accounting context: never over-interpret into a narrative or a literary sentence; a fragment remains an excerpt from a register. (9) The chronology of the Knossos tablets is debated. (10) Ambiguity of the ligatures and monograms of logograms. Golden rule: sign-by-sign transliteration first, interpretation afterward, uncertainty always made explicit.

Decipherment history

The decipherment of Linear B is an emblematic achievement. The tablets were brought to light as early as 1900 at Knossos by Arthur Evans, who distinguished Linear A from Linear B but long kept the corpus largely unpublished and believed the language to be non-Greek ("Minoan"). Carl Blegen's excavations at Pylos (1939) yielded a vast mainland lot. The American epigraphist Emmett L. Bennett Jr. established the reliable sign inventory, a numbering of the signs, and the reference editions of the Pylos tablets, founding the discipline. The American Alice Kober (in the 1940s) demonstrated through rigorous analysis that the language was inflectional and constructed series (the "triplets," in Ventris's phrase) revealing declensional paradigms, without, however, assigning sound values. The English architect Michael Ventris, building on the work of Kober and Bennett, constructed a consonant/vowel "grid." In 1952, through a series of inspired hypotheses (the identification of Cretan toponyms such as ko-no-so = Knossos, a-mi-ni-so = Amnisos), he found that the values yielded Greek — a conclusion he had not expected. The philologist John Chadwick joined him; together they published in 1953 the foundational article "Evidence for Greek Dialect in the Mycenaean Archives" (Journal of Hellenic Studies 73, 1953, pp. 84–103), then in 1956 the work Documents in Mycenaean Greek. The "tripod tablet" of Pylos (PY Ta 641), discovered by Blegen in 1952 — who immediately recognized ti-ri-po-de as the Greek word for "tripods" — and then read with Ventris's values (the concordance between vessel logograms and adjectives: ti-ri-po "tripod," qe-to-ro-we "four-handled"), furnished in 1953 the decisive confirmation. Ventris died in a car accident in 1956.

Common formulae & expressions

qe

"and" (enclitic conjunction = Greek τε, kʷe), suffixed to the coordinated word

e-ke

"he has / holds / possesses" (ekhei), a very frequent verb in landholding records

o-pe-ro / o-pe-ro-sa

"deficit, lacking, owed" (ophelos/opheronsa), notes an arrears or an unfulfilled obligation

to-so / to-sa

"so much, in total so many" (tossos/tossa), introduces or closes a total (masc./fem.-neut.)

pa-ro

"from, in the possession of, at the house of" (paro), a frequent administrative preposition

wa-na-ka / wa-na-ka-te-ro

"king (wanax)" / "royal, of the king" (adjective wanakteros)

ra-wa-ke-ta

"leader of the people/army" (lawagetas), the second dignitary after the wanax

do-e-ro / do-e-ra

"slave, servant" masc./fem. (dowelos/dowela), often "slave of a god"

ko-wo / ko-wa

"boy, son" / "girl" (korwos/korwa), frequent in personnel lists

di-we / po-se-da-o-ne

"to Zeus" (dative Diwei) / "to Poseidon" (dative), headings of religious offerings

o- (initial)

prefix/particle often opening a rubric or the heading of a tablet (interpretation debated)

te-o-jo do-e-ro

"slave of the god" (theoio dowelos), a category of personnel dependent on the sanctuary

Corpora & reference editions

• CoMIK — Corpus of Mycenaean Inscriptions from Knossos (Chadwick, Godart, Killen, Olivier, Sacconi, Sakellarakis, 4 vols., 1986–1998)
• The Pylos tablets: The Pylos Tablets (E. L. Bennett Jr., 1955); The Pylos Tablets Transcribed I–II (Bennett & Olivier, 1973/1976)
• TITHEMY / the corpus of Thebes, Mycenae, Tiryns; editions of the Theban tablets (Aravantinos, Godart, Sacconi)
• DĀMOS — Database of Mycenaean at Oslo (online database of the tablets)
• LiBER — Linear B Electronic Resources (CNR, Italy)
• GORILA — Recueil des inscriptions en Linéaire A (Godart & Olivier, 1976–1985), for the Minoan mother-script (for comparison)

Bibliographic references

• Michael Ventris & John Chadwick, "Evidence for Greek Dialect in the Mycenaean Archives," Journal of Hellenic Studies 73, 1953, pp. 84–103 — Foundational article announcing the decipherment
• Michael Ventris & John Chadwick, Documents in Mycenaean Greek, Cambridge University Press, 1956 (2nd ed. by Chadwick, 1973) — The reference work: script, language, society, 300 annotated texts
• John Chadwick, The Decipherment of Linear B, Cambridge University Press, 1958 — A classic and accessible account of the history of the decipherment
• John Chadwick, The Mycenaean World, Cambridge University Press, 1976 — A synthesis of Mycenaean society based on the tablets
• Emmett L. Bennett Jr., The Pylos Tablets: Texts of the Inscriptions Found 1939–1954, Princeton University Press, 1955 — The foundational edition of Mycenaean epigraphy; establishment of the sign inventory and numbering
• Yves Duhoux & Anna Morpurgo Davies (eds.), A Companion to Linear B: Mycenaean Greek Texts and their World, 3 vols., Peeters (Louvain-la-Neuve), 2008 / 2011 / 2014 — The contemporary scholarly reference handbook
• Antonín Bartoněk, Handbuch des mykenischen Griechisch, Carl Winter (Heidelberg), 2003 — The complete reference grammar of Mycenaean Greek
• Andrew L. Sihler, New Comparative Grammar of Greek and Latin, Oxford University Press, 1995 — A comparative Indo-European framework incorporating the Mycenaean data