Iberian

Overview

Iberian (endonym unknown) is the best-attested Palaeohispanic language, spoken along the Mediterranean seaboard of the Iberian Peninsula, from southern France (Languedoc-Roussillon) to eastern Andalusia, between roughly the 6th/5th century BC and the 1st century AD. It is documented by an epigraphic corpus of some two thousand inscriptions (estimates commonly cited between ~2000 and ~2300 depending on the counting method): lead, stone, ceramic, coins, mosaics. Its writing system — the Iberian script, principally in its north-eastern (Levantine) variant — is LARGELY DECIPHERED as regards the phonetic value of the signs (we can read, that is, vocalize, most texts). The LANGUAGE itself, however, remains POORLY UNDERSTOOD: neither its genetic family nor the translation of the majority of its lexicon has been established. Two distinct planes must therefore be clearly separated: (1) the decipherment of the script, which is very advanced; (2) the understanding of the language, which is partial and limited to onomastics, a few grammatical elements, and the numeral system. The position generally held by scholarship is that Iberian is NOT an Indo-European language and that no genetic relationship with Basque has been demonstrated, despite striking resemblances (numerals, onomastics) that are most often explained by contact (Sprachbund) rather than by a proven common origin — even though a hypothesis of ancient kinship (Basco-Iberism) continues to be debated.

History of the people

The "Iberians" designate, in the Greek and Latin sources (Hecataeus, Herodotus, Strabo, Pliny, Avienus), a set of peoples of the Mediterranean seaboard of the Iberian Peninsula and of Languedoc, comprising distinct entities: Ausetani, Ilergetes, Laietani, Cessetani, Edetani, Contestani, Bastetani, among others. This was not a unified nation but a cultural and linguistic area sharing material traits (fortified urbanism/oppida, painted ceramics, sculpture — e.g. the Lady of Elche —, coinage) and the use of the Iberian script. Their Iron Age culture developed in intense contact with the Phoenician and Greek colonies (Emporion/Ampurias, Massalia) and later with Carthage and Rome. Their economy was based on agriculture, metallurgy, and Mediterranean trade. Politically fragmented into cities and confederations, they played a role in the Punic Wars and were then integrated into Roman Hispania. Romanization progressively dissolved Iberian identity and language between the 2nd century BC and the 1st century AD. The Iberians must be distinguished from the Celtiberians (central peninsula, Celtic language) and from the populations of the south-west (Tartessos).

History of the language

Iberian is attested from roughly the 6th/5th century BC (earliest inscriptions) to the 1st century AD. It was the language of a vast Mediterranean coastal zone of the peninsula and of southern Gaul, probably also serving as a contact/trade language between diverse populations (which complicates the question of whether all "Iberians" spoke it natively). Its script is adapted from earlier models (Phoenician and Greek, via the scripts of the south-west of the peninsula). Roman expansion from the end of the 3rd century BC (Second Punic War) brought about a progressive Romanization; Iberian inscriptions became rarer in the 1st century BC and disappeared in the 1st century AD, the language being supplanted by Latin. No linguistic descent is proven: Iberian died out with no identified continuator (Basque, the surviving non-Indo-European language of the region, is NOT its demonstrated heir — at most a possible typological relative or contact partner, UNCERTAIN). Genetic status: an ISOLATE / unclassified language, non-Indo-European according to the consensus.

The script

Iberian is written in several distinct graphic systems, all local except the Greco-Iberian alphabet. (1) NORTH-EASTERN SCRIPT (so-called Levantine Iberian), by far the most frequent (the majority of the corpus), used in Catalonia, Aragon, Valencia, and southern France; it is today the object of the most secure decipherment. (2) SOUTH-EASTERN SCRIPT (Meridional), typologically related but derived more directly from the south-western models, less well read and less abundant. (3) GRECO-IBERIAN ALPHABET, an adaptation of the Ionic Greek alphabet used mainly in the region of Alicante/Contestania, wholly alphabetic and therefore legible without ambiguity (invaluable for calibrating the values of the other systems). A major typological trait common to all (except Greco-Iberian): these are SEMI-SYLLABARIES, in which the stops are written with SYLLABIC signs (one consonant+vowel value per sign, e.g. ba, be, bi, bo, bu), while vowels, sibilants, nasals, and liquids are written with ALPHABETIC (monophonemic) signs. The direction of writing is generally left to right for the north-eastern script. The north-eastern script exists in two states: the SIMPLE (non-dual) system and the DUAL SYSTEM, in which certain occlusive syllabic signs (dentals and velars) present a variant marked with an additional stroke distinguishing voiceless from voiced — a distinction that the simple system did not note. NOTE the convention: according to the reference analysis (Maluquer de Motes, then Ferrer i Jané), it is the SIMPLE/unmarked sign that notes the VOICED stop (d, g) and the COMPLEX sign, marked with the additional stroke, that notes the VOICELESS stop (t, k) — and not the reverse. The dual is almost exclusive to the older inscriptions (chiefly 4th–3rd c. BC).

Sign system

The non-dual north-eastern script comprises about 28 signs: 5 VOWELS (a, e, i, o, u); 15 SYLLABIC SIGNS for the stops, organized into three series × five vowels — labial (ba, be, bi, bo, bu), dental (ta, te, ti, to, tu), and velar (ka, ke, ki, ko, ku); and 8 monophonemic CONSONANTAL SIGNS: one lateral (l), two vibrants/rhotics (transcribed r and ŕ), two sibilants (transcribed s and ś), and nasals (n, m, plus a sign sometimes transcribed ḿ depending on the author). TRANSCRIPTION CONVENTIONS (Untermann's, widely followed): the two rhotics and the two sibilants are distinguished by the accent (r/ŕ, s/ś); the dental/velar stops are noted as neutral (t, k) in the simple system for want of a voiceless/voiced distinction. In the DUAL, the UNMARKED variant notes the voiced stop (d, g) and the variant MARKED with an additional stroke notes the voiceless stop (t, k). The labial series has only one sign per vowel (transcribed b), the p/b distinction not being systematically marked. IMPORTANT: the conventional transcription yields a reliable VOCALIZATION, but it does NOT give the meaning: to read Iberian is to pronounce it, not to understand it. The decipherment of the values rests historically on coin legends (place-names known from the Greco-Latin sources and from bilingual coins) and on Latin onomastic documents such as the Bronze of Ascoli.

Phonology

Phonology reconstructed cautiously from the sign values and from Greek/Latin renderings of Iberian names. VOWELS: a five-quality system /a e i o u/, with the probable existence of diphthongs (ai, ei, au). No solid evidence of a length distinction. CONSONANTS: two distinct sibilants (s vs ś), whose exact nature — apical/laminal or affricated — remains debated (uncertain). Two rhotics (r vs ŕ), a real opposition but with uncertain phonetic realization (trill vs tap?). Nasals m, n and possibly a third nasal unit. Lateral l. Stops at the three places of articulation (labial, dental, velar); the voicing opposition (voiceless/voiced) is noted in the dual system but graphically neutralized in the simple. Notable phonotactic constraints inferred from onomastics: rarity of complex consonant clusters, apparent absence of frequent initial /p/, syllable structure tending toward (C)V. Probable absence of f and of certain fricatives. These traits (5 vowels, two r's, two s's, a wealth of nasals) are typologically close to ancient Basque, which has fed the contact hypothesis; but any fine phonetic reconstruction beyond the vocalization of the signs remains UNCERTAIN.

Grammar — essentials

The grammar is known only very fragmentarily. RELATIVELY SECURE ACQUISITIONS: (1) a well-identified two-member onomastics (personal names composed of two formants drawn from a recurring stock: iltiŕ/iltur, bels, taŕ, iltun, śosin, atin, etc.); (2) a vigesimal NUMERAL system (see formulae); (3) a few suffixes probably case/relational, added to anthroponyms (e.g. -ar, -en, often interpreted as genitive/possessive "of X"), as well as elements such as -te, -ku whose function is disputed. The verbal system, the overall syntax, and the inflectional morphology remain largely UNKNOWN or hypothetical. No reliable continuous translation of a non-onomastic text is possible. Any more far-reaching grammatical description belongs to the realm of working hypotheses (e.g. "pragmatic" or agglutinative readings proposed by typological analogy with Basque) and must be flagged as such.

Reading tips

Reading method and pitfalls for honest translation assistance. (1) FIRST IDENTIFY THE WRITING SYSTEM (north-eastern simple vs dual vs south-eastern vs Greco-Iberian): in the dual system, the mark (additional stroke) notes the VOICELESS stop (t, k) and the simple sign the VOICED stop (d, g); the simple system makes no distinction — never force an absent distinction in the simple system, nor invert the dual convention. (2) TRANSLITERATE sign by sign according to Untermann's conventional values, respecting the diacritics (ŕ, ś, possibly ḿ) — they are phonemically distinctive. (3) SEGMENT: look for word boundaries (sometimes marked by dots/interpuncts), isolating first the ONOMASTICS (two-member names with known formants) and the NUMERALS, the only relatively secure domains. (4) DO NOT TRANSLATE the rest as if the meaning were known: for the majority of the lexicon, the semantic value is unknown. Provide a TRANSLITERATION + vocalization, flag the identifiable segments (probable proper name, numeral, possible case suffix), and mark all the rest as untranslated. (5) BEWARE the Basque temptation: the numeral correspondences and a few formants are real but do NOT constitute a general translation key; any lexical "translation" via Basque beyond the numerals and a few rare terms is speculative and must be flagged as such. (6) Always refer back to the reference editions (MLH, Hesperia database) for the reading of contested signs.

Pitfalls

MAJOR PITFALLS. (1) Confusing "to read" and "to understand": Iberian can be vocalized, but semantic translation remains largely impossible — never present a vocalization as a translation. (2) The Basco-Iberian myth: the resemblances (numerals, a few onomastic formants) are real, but GENETIC kinship is NOT demonstrated; using Basque as a dictionary to translate Iberian beyond the numerals is a widespread methodological error to be avoided. (3) Dual vs simple system, and the sense of the mark: in the dual, it is the MARKED sign (additional stroke) that notes the VOICELESS stop (t, k) and the simple sign the VOICED stop (d, g); inverting this convention, or applying a voicing distinction to a text in the simple system, distorts the reading. (4) Distinctive diacritics: neglecting r/ŕ and s/ś loses real phonemic oppositions. (5) Segmentation: mis-dividing the words (interpuncts absent or irregular) leads to false units; always give priority to the known onomastics. (6) Confusing the three scripts (north-eastern, south-eastern, Greco-Iberian) and confusing them with Celtiberian (an Indo-European language written in a related script) or with Tartessian/south-western — these are different systems/languages. (7) Over-interpreting the "formulae": eban, aŕe take, śalir have an identified functional context but a precise meaning that is debated.

Decipherment history

The decipherment of the WRITING (not of the language) is the founding work of Manuel Gómez-Moreno, who around 1922 established the SEMI-SYLLABIC nature of the north-eastern script and assigned the values of the occlusive signs, relying on the many coin legends referring to place-names known from the Greco-Latin sources, as well as on Latin onomastic lists (Bronze of Ascoli). His results, published and developed in the following decades, were at first received with skepticism and then confirmed and refined. In the 20th century, Jürgen Untermann (1928–2013) systematized the corpus and the onomastics in the monumental Monumenta Linguarum Hispanicarum (the series begun in the 1960s, first volume 1975, last volumes posthumous 2018–2019). Untermann's analysis of the two-member onomastics was a major milestone for segmenting the texts. An important advance was the recognition of the DUAL SYSTEM (marked variants distinguishing voiceless from voiced), identified by Joan Maluquer de Motes (1968) and subsequently developed by later research, notably Joan Ferrer i Jané (from 2005). Finally, the reconstruction of the NUMERAL SYSTEM (vigesimal, very close to Basque) was initiated by Eduardo Orduña (2005) and developed by Joan Ferrer i Jané, constituting the principal recent success in lexical understanding. DESPITE all this, the language remains not "deciphered" in the semantic sense: one reads without understanding.

State of research

The state of "decipherment" must be carefully separated into two planes. GRAPHIC PLANE (writing): very advanced — the values of the signs of the north-eastern script (simple and dual) and of Greco-Iberian are known with a good degree of reliability; the great majority of texts can be transliterated and vocalized. The south-eastern (Meridional) script is less securely read (several signs of uncertain or debated value). LINGUISTIC PLANE (language): weakly understood — solid acquisitions limited to: (a) the vigesimal numeral system; (b) the structure of the two-member onomastics and a stock of formants; (c) a few probable case suffixes; (d) the contextual identification of certain funerary/commercial formulae. What REMAINS UNKNOWN: the genetic affiliation of the language, the verbal system, the overall syntax, and the semantics of nearly the whole vocabulary. OBSTACLES: the absence of an extended bilingual (no "Rosetta"); a corpus made up mostly of short, repetitive, onomastic, or fragmentary texts; the possibly composite/contact character of the language. Notable but inconclusive ATTEMPTS: Basco-Iberian comparisons beyond the numerals, and various etymological proposals that have remained speculative.

Common formulae & expressions

ban / bi / laur / borste / abaŕ / oŕkei

Numerals: 'one', 'two', 'four', 'five', 'ten', 'twenty' (vigesimal system, very close to Basque bat/bi/lau/bost/hamar/hogei); erder ≈ 'half'. The numerals combine (e.g. oŕkei + borste ≈ 25, cf. Basque hogei ta bost) and appear in weight/measure contexts. Numeration = one of the few lexical domains held to be understood (Orduña 2005, Ferrer i Jané).

iltiŕ- / ilti- (e.g. iltiŕta, place-name/formant)

A very frequent onomastic and toponymic element, often associated with the notion of 'town/city' (hypothesis); the base of numerous place-names and coin legends (e.g. the coinage of Iltirta, identified with Ilerda/Lleida).

eban / teban / ebanen

A recurring term on funerary stelae. The dominant interpretation (Untermann, Velaza, based on a bilingual from Saguntum): the equivalent of Latin coeravit, marking the dedicant/the one who had the monument erected. An alternative interpretation as a filiation marker ('filius') is defended notably by Rodríguez Ramos (2001). PLAUSIBLE but still debated.

aŕe take (or aŕe teike)

A frequent formula at the head of funerary stelae, often likened to an opening formula of the type 'here lies' / 'hic est'; the exact meaning is UNCERTAIN, but its identification as a funerary formula is well established by context.

śalir / śalirnai

A recurring term on lead tablets and supports of a commercial character, often interpreted as relating to money/value/payment ('money', 'unit of value'?); a contextual interpretation, UNCERTAIN.

-ar / -en (suffixed to an anthroponym)

Suffixes probably case/relational (genitive/possessive: 'of X'?), frequent on objects of ownership and on dedications; a plausible function, the precise value debated.

Corpora & reference editions

• Monumenta Linguarum Hispanicarum (MLH), ed. J. Untermann, Reichert Verlag, Wiesbaden (series ~1975-2019); the standard reference for Palaeohispanic inscriptions. Vol. III (Iberian inscriptions, 1990), Vol. V.2 = Lexikon der iberischen Inschriften (2019), Vol. VI (toponymy, 2018)
• Hesperia database (Banco de Datos Hesperia de Lenguas Paleohispánicas, hesperia.ucm.es), online — the current reference digital epigraphic corpus
• Lexikon der iberischen Inschriften (MLH V.2), N. Moncunill & J. Velaza, 2019 — an annotated glossary of the Iberian lexicon
• Palaeohispanica (journal, IFC Zaragoza) — publications of new inscriptions and studies
• Earlier historical corpus: Monumenta Linguae Ibericae, E. Hübner, 1893 (superseded by MLH)

Bibliographic references

• Jürgen Untermann, Monumenta Linguarum Hispanicarum, Reichert, ~1975-2019 — The reference corpus of Palaeohispanic inscriptions and of Iberian onomastics; the basis of all serious research. Existence and dates verified.
• Noemí Moncunill & Javier Velaza, Lexikon der iberischen Inschriften (MLH V.2), 2019 — A reasoned glossary of the attested Iberian lexicon; the state of the art on lexical segments and their cautious interpretation. Verified.
• Jesús Rodríguez Ramos, Análisis de epigrafía íbera, Anejos de Veleia, Universidad del País Vasco, 2004 — A synthesis on the script and the reading of Iberian inscriptions. Existence verified (ISBN 84-8373-678-0).
• Joan Ferrer i Jané, works on the dual system and the numerals (Palaeohispanica, from 2005) — Major recent contributions on the dual system (marked variant = voiceless) and on Iberian numeration. Verified.
• Eduardo Orduña, «Los numerales ibéricos y el protovasco» and related works, from 2005 — The origin of the identification of the Iberian vigesimal numeral system and of its typological comparison with Basque. Verified.
• Manuel Gómez-Moreno, decipherment of the Iberian script (from ~1922) — The founding decipherment of the sign values of the north-eastern semi-syllabary. Date ~1922 verified.