Etruscan language

The Etruscan language was spoken and written by the Etruscan civilization in the ancient region of Etruria (modern Tuscany plus western Umbria and northern Latium) and in parts of Lombardy, Veneto, and Emilia-Romagna (where the Etruscans were displaced by Gauls), in Italy. However, Latin superseded Etruscan completely, leaving only a few documents and a few loanwords in Latin e.g., persona from Etruscan ersu, and some place-names, such as Roma.

History of Etruscan literacy
Etruscan literacy was widespread over the Mediterranean shores, as can be seen by about 13,000 inscriptions (dedications, epitaphs etc), most fairly short, but some of some length. They date from about 700 BC.

The Etruscans had a rich literature, as noted by Latin authors. Unfortunately only one book (now unreadable) has survived, although there is always some possibility that more will turn up. By AD 100, Etruscan had been replaced by Latin.

Only a few educated Romans with antiquarian interests, such as Varro, could read Etruscan. The last person known to have been able to read it was the Roman emperor Claudius (10 BC – AD 54), who &mdash; in the context of his work in twenty books about the Etruscans, Tyrrenikà (now lost) &mdash; compiled a dictionary (also lost) by interviewing the last few elderly rustics who still spoke the language. Urgulanilla, his first wife, was Etruscan.

Livy and Cicero were both aware that highly specialized Etruscan religious rites were codified in several sets of books written in Etruscan under the generic Latin title Etrusca Disciplina. The Libri Haruspicini dealt with divination from the entrails of the sacrificed animal, the Libri Fulgurales expounded the art of divination by observing lightning. A third set, the Libri Rituales, would have provided us with the key to Etruscan civilization: its wider scope embraced Etruscan standards of social and political life as well as ritual practices. According to the 4th century Latin writer Servius, a fourth set of Etruscan books existed, dealing with animal gods, but it is probably unlikely that any contemporary scholar could have read Etruscan at such a late date. Christian authorities collected such works of paganism and destroyed them during the 5th century; the single surviving Etruscan book, Liber Linteus, being written on linen, survived only by being used as mummy wrappings.

Etruscan had some influence over Latin. A few dozen words were borrowed by the Romans and some of them can be found in modern languages.

Geographic distribution
Inscriptions have been found in north-west and west-central Italy, in the region that even now bears the name of the Etruscans, Tuscany (from Latin tusc "Etruscans"), as well as in today's Latium north of Rome, in today's Umbria west of the Tiber, around Capua in Campania and in the Po valley to the north of Etruria. Presumably this range is a maximum Italian homeland where the language was at one time spoken.

Outside of Italy inscriptions have been found in Africa, Corsica, Elba, Gallia Narbonensis, Greece, the Balkans and the Black Sea. By far the greatest concentration is in Italy.

An inscription found on Lemnos in 1886, is in an alphabet practically identical to that of Etruscan.

Classification
The Etruscan language has been difficult to analyze, which is attributable to its being an isolate. The phonology is known through the alternation of Greek and Etruscan letters in some inscriptions (for example, the Iguvine Tables), and many individual words are known through loans into or from Greek and Latin, as well as explanations of Etruscan words by ancient authors. A few concepts of word formation have been formulated (see below). Knowledge of the language is incomplete.

In the 1st century BC the Greek historian Dionysius of Halicarnassus stated that the Etruscan language was unlike any other. Bonfante, a leading scholar in the field, says "... it resembles no other language in Europe or elsewhere ...."

Tyrsenian family
The majority consensus is that Etruscan is related only to other members of what is called the Tyrsenian language family which in itself is an isolate family, that is, unrelated to other language groups by any known relationship. Since Rix (1998) it is widely accepted that Tyrsenian is composed of Rhaetic and Lemnian together with Etruscan.

Another possible Aegean language related with Etruscan is Minoan. The idea of a relation between the language of the Aegean Linear scripts was taken into consideration as the main hypothesis by Michael Ventris before discovering that in fact the language behind the more modern Linear B script was Mycenean, a  Greek dialect. Facchetti, a researcher who has dealt with both languages (Etruscan and Minoan) has put forward again this hypothesis, comparing some of the Minoan words of known meaning with some similar Etruscan words Some modern scholars assert that the Tyrsenian languages are distantly related to the Indo-European family. More specifically Frederik Woudhuizen suggests a relation to the Anatolian branch of the family. Woudhuizen revived a conjecture to the effect that the Tyrsenians came from Anatolia, including Lydia, whence they were driven out by the Cimmerians in the early Iron Age, 750-675 BC, leaving some colonists on Lemnos. He makes a number of comparisons of Etruscan to Luwian and asserts that Etruscan is modified Luwian. He accounts for the non-Luwian features as a Mysian influence: "deviations from Luwian ... may plausibly be ascribed to the dialect of the indigenous population of Mysia." According to Woudhuizen, the Etruscans were colonizing the Latins. The Etruscans brought the alphabet from Anatolia.

More recently Robert S.P. Beekes presented a similar case, but argued that the people later known as the Lydians and Etruscans had originally lived in north west Anatolia, with a coastline to the Sea of Marmara, whence they were driven by the Phrygians c. 1,200 BC, leaving a remnant known in antiquity as the Tyrsenoi. A segment of this people moved south-west to Lydia, becoming known as the Lydians, while others sailed away to take refuge in Italy, where they became known as Etruscans. The Etruscan language could therefore have been related to a non-Indo-European substratum of Lydian.

Both of these accounts draw on the story by Herodotus (i, 94) of the Lydian origin of the Etruscans. Dionysius of Halicarnassus (book 1) rejected this account of the people he called the Tyrrhenians, partly on the authority of Xanthus, a Lydian historian, who had no knowledge of the story, and partly on what he judged to be the different languages, laws and religions of the two peoples.

Other
Historically, a relation of Etruscan to the Semitic languages has also been discussed. This suggestion has been obsolete since the later 19th century. Various other speculative proposals have been made in the 19th and 20th centuries, all of them failing to gain significant acceptance.

The interest in Etruscan antiquities and the mysterious Etruscan language found its modern origin in a book by a Dominican friar, Annio da Viterbo (d. 1502), also known as Il Pastura, the cabalist and orientalist, now remembered mainly for literary forgeries. He guided Pinturicchio's allegorical frescoes for Pope Alexander VI's Vatican apartments. In 1498 Annio published his antiquarian miscellany titled Antiquitatum variarum (in 17 volumes) where he put together a fantastic theory in which both the Hebrew and Etruscan languages were said to originate from a single source, the "Aramaic" spoken by Noah and his descendants, founders of Etruscan Viterbo. Annio also started to excavate Etruscan tombs, unearthing sarcophagi and inscriptions, and made a bold attempt at deciphering the Etruscan language. The theory of the Semitic origins still found its supporters in 19th century scholarship. In 1858, the last attempt was made by Johann Gustav Stickel, Jena University: "Das Etruskische (...) als Semitische Sprache erwiesen". A reviewer concluded that Stickel brought forward every possible argument which would speak for that hypothesis, but he proved the opposite of what he had attempted to do (Johannes Gildemeister in ZDMG 13, 1859, 289-304).

In 1861 Robert Ellis proposed that Etruscan was related to Armenian. Some scholars also see in Urartian art, architecture, language and general culture traces of kinship to the Etruscans of the Italian peninsula.

Relation with Albanian in particular has been advanced by a number of people, notably Zacharie Mayani, as well as earlier writers such as Ascoli, 1877, E. Schneider, 1889, Thomopoulos, 1912, Buonamici, 1919.

Another proposal, currently pursued mainly by a few linguists from the former Soviet Union, suggests a relationship with Northeast Caucasian (or Nakh-Daghestanian) languages.

Alphabet
The Latin alphabet that is used in English owes its existence to the Etruscan writing system, which was adapted for Latin in the form of the Old Italic alphabet. The Etruscan alphabet employs a Euboean variant of the Greek alphabet using the letter digamma and was in all probability transmitted through Pithecusae and Cumae, two Euboean settlements in southern Italy. This system is ultimately derived from West Semitic scripts.

The Etruscans recognized a full 26-letter alphabet, which they depicted as itself for decoration on some objects such as an occasional ink-jar; for example, the "rooster ink-stand." This has been termed the model alphabet. They did not use four letters of it, mainly because Etruscan had no voiced stops, b, d and g, and also no o. They innovated one letter for f.

Text
Writing was from right to left except in archaic inscriptions, which might use boustrophedon. An example found at Cerveteri used left to right. In the earliest inscriptions, the words are continuous. From the 6th century BC, they are separated by a dot or a colon, which might also separate syllables. Writing was phonetic; the letters represented the sounds and not conventional spellings. On the other hand, many inscriptions are highly abbreviated and often casually formed, so the identification of many individual letters is sometimes difficult. Spelling might vary from city to city, probably reflecting differences of pronunciation.

Impossible consonants
Speech featured a heavy stress on the first syllable of a word, causing syncopation by weakening of the remaining vowels, which then were not represented in writing: Alcsntre for Alexandros, Rasna for Rasena. This speech habit is one explanation of the Etruscan "impossible consonant clusters." The resonants however may have been syllabic, accounting for some of the clusters (see below under Consonants). In other cases the scribe sometimes inserted a vowel: Greek Herakles became Hercle by syncopation and then was expanded to Herecele. Pallottino regarded this variation in vowels as "instability in the quality of vowels" and accounted for the second phase (e.g. Herecele) as "vowel harmony, i.e., of the assimilation of vowels in neighboring syllables ...."

Phases
The writing system had two historical phases: the archaic, 7th to 5th century BC, which used the early Greek alphabet, and the later, 4th to 1st century BC, which modified some of the letters. In the later period syncopation increased.

The alphabet went on in modified form after the language disappeared. In addition to being the source of the Roman alphabet, it has been suggested that it passed northward into Venetic and from there through Raetia into the Germanic lands, where it became the Futhark, a system of runes.

Bilinguals
The Pyrgi Tablets are a bilingual text in Etruscan and Phoenician engraved on three gold leaves, one for the Phoenician and two for the Etruscan. The Etruscan is in 16 lines, 37 words. The date is roughly 500 BC.

The tablets were found in 1964 by Massimo Pallottino during an excavation 20 miles from the city of Rome, at the ancient Etruscan port of Pyrgi.

Longer texts
According to Rix and his collaborators only two unified (though fragmentary) texts are available in Etruscan: Some additional longer texts are:
 * The Liber Linteus used for mummy wrappings (at Zagreb, Croatia). Roughly 1200 words of readable text, mainly repetitious prayers yielding about 50 lexical items.
 * The Tabula Capuana (the inscribed tile from Capua). About 300 readable words in 62 lines, dating to the 5th century BC.
 * The lead foils of Punta della Vipera, about 40 legible words having to do with ritual formulae. Dated to about 500 BC.
 * The Cippus Perusinus, a stone slab (cippus) found at Perugia. Contains 46 lines, 130 words.
 * The Tabula Cortonensis, a bronze tablet from Cortona recording a legal contract. About 200 words.
 * The Piacenza Liver, a bronze model of a sheep's liver representing the sky, with the engraved names of the gods ruling different sections.

Inscriptions on monuments
The main material repository of Etruscan civilization is or was its tombs. Public and private buildings were dismantled and the stone reused centuries ago. The tombs remain as they were except for the ravages of time and the activities of plunderers. More tombs continue to be found regularly.

The tombs are the main source of portables in collections throughout the world, provenance unknown. The Etruscans lived well and valued art. Their objets d'art are of incalculable value, causing a brisk black market and equally brisk law enforcement effort. It is against the law to remove objects from Etruscan tombs unless authorized by the Italian government.

The total number of tombs is unknown due to the magnitude of the task of cataloguing them. They are of many different types. Especially fruitful are the hypogeal or "underground" chamber or system of chambers cut into tuff and covered by a tumulus. The interior of the tomb represents a habitation of the living stocked with furniture and favorite objects. The walls may display painted murals, the predecessor of wallpaper. Tombs are identified as Etruscan dating form the Villanovan period to about 100 BC, when presumably the cemeteries were abandoned in favor of Roman ones. Some of the major cemeteries are as follows:
 * Caere or Cerveteri, a UNESCO site. Three complete necropolises with streets and squares. Many hypogea are concealed beneath tumuli retained by walls; others are cut into cliffs. The Banditaccia necropolis contains more than 1000 tumuli. Access is through a door.


 * Tarquinia, Tarquinii or Corneto, a UNESCO site. Approximately 6000 graves dating from the Villanovan (9th & 8th centuries BC) distributed in necropolises, the main one being the Monterozzi hypogea of the 6th - 4th centuries BC. About 200 painted tombs display murals of various scenes with call-outs and descriptions in Etruscan. Elaborately carved sarcophagi of marble, alabaster and nenfro include identificatory and achievemental inscriptions. The Tomb of Orcus at the Scatolini necropolis depicts scenes of the Spurinna family with call-outs.


 * Inner walls and doors of tombs and sarcophagi.
 * Engraved steles (tombstones)
 * ossuaries

Votives
Votive gifts

Specula
A speculum is a circular or oval hand-mirror used predominantly by Etruscan women. Speculum is Latin; the Etruscan word is malena or malstria. Specula were cast in bronze as one piece or with a tang into which a wooden, bone or ivory handle fitted. The reflecting surface was created by polishing the flat side. A higher percentage of tin in the mirror improved its ability to reflect. The other side was convex and featured intaglio or cameo scenes from mythology. The piece was generally ornate.

About 2300 specula are known from collections all over the world. As they were popular plunderables, the provenance of only a minority is known. An estimated time window is 530-100 BC. Most probably came from tombs.

Many bear inscriptions naming the persons depicted in the scenes, for which reason they are often called picture bilinguals. In 1979, Massimo Pallottino, then president of the Istituto di Studi Etruschi ed Italici initiated the Committee of the Corpus Speculorum Etruscanorum (CSE), which resolved to publish all the specula and set editorial standards for doing so.

Since then the committee has grown, acquiring local committees and representatives from most institutions owning Etruscan mirror collections. Each collection is published in its own fascicle by diverse Etruscan scholars.

Cistae
A cista is a bronze container of circular, ovoid or more rarely rectangular shape used by women for the storage of sundries. They are ornate, often with feet and lids to which figurines may be attached. The internal and external surfaces bear carefully crafted scenes usually from mythology, usually intaglio, rarely part intaglio, part cameo.

Cistae date from the Roman Republic of the 4th and 3rd centuries BC in Etruscan contexts. They may bear various short inscriptions concerning the manufacturer or owner or subject matter. The writing may be Latin, Etruscan or both.

Excavations at Praeneste, an Etruscan city turned Roman, turned up about 118 cistae, one of which has been termed "the Praeneste cista" or "the Ficoroni cista" by art analysts, with special reference to the one manufactured by Novios Plutius and given by Dindia Macolnia to her daughter, as the archaic Latin inscription says. All of them are more accurately termed "the Praenestine cistae."

Rings and ringstones
Among the most plunderable portables from the Etruscan tombs of Etruria are the finely engraved gemstones set in patterned gold to form circular or ovoid pieces intended to go on finger rings. Of the magnitude of one centimeter, they are dated to the Etruscan floruit from the 2nd half of the 6th to the 1st centuries BC. The two main theories of manufacture are native Etruscan and Greek.

The materials are mainly dark red cornelian with agate and sard coming in from the 3rd to the 1st centuries BC along with purely gold finger rings of a hollow engraved bezel. The engravings, mainly cameo, but sometimes intaglio, depict scarabs at first and then scenes from Greek mythology, often with heroic personages called out in Etruscan. The gold setting of the bezel bears a border design, such as cabling.

Coins
Etruscan-minted coins date ca. 500-200 BC. Use of the Euboïc-Syracusan standard, based on the silver litra of 13.5 grams maximum, indicates the custom, like the alphabet, came from Greece. Roman coinage supplanted Etruscan, but the basic Roman coin, the sesterce, is believed to have been based on the 2.5 denomination Etruscan coin. Etruscan coins have turned up in caches or individually in tombs and in excavations seemingly at random, concentrated, of course, in Etruria.

Etruscan coins were in gold, silver and bronze, the gold and silver usually having been struck on one side only. The coin bore a denomination, a minting authority name, and a cameo motif. Gold denominations were in units of silver; silver, in units of bronze. Full or abbreviated names are mainly pupluna (Populonia), Vatl or Veltuna (Vetulonia), Velathri (Volaterrae), Velzu or Velznani (Volsinii) and Cha for Chamars (Camars). Insignia are mainly heads of mythological characters or depictions of mythological beasts arranged in a symbolic motif: Apollo, Zeus, Janus, Athena, Hermes, griffin, gorgon, sphinx, hippocamp, bull, snake, eagle, etc.

Recent discoveries
A book of gold sheets bound with gold rings went on display in May 2003 at the National History Museum in Sofia, Bulgaria. It consists of six bound sheets of 24-carat (100%) gold, with low-reliefs of a horseman, a mermaid, a harp and soldiers, with text. It was claimed to have been discovered about 1940 in a tomb uncovered during digging for a canal along the Strouma river in south-western Bulgaria, kept secretly and anonymously donated by its 87-year-old owner, living in the Republic of Macedonia. While officials of the museum have claimed that the text of the book is in Etruscan, they have not published the text, and therefore specialists in Etruscan have been unable to verify their claim.

Sounds
In the tables below, conventional letters used for transliterating Etruscan are accompanied by likely pronunciation in IPA symbols within the square brackets, followed by examples of the early Etruscan alphabet which would have corresponded to these sounds:

Vowels
The Etruscan vowel system consisted of four distinct vowels. Vowels "o" and "u" appear to have not been phonetically distinguished based on the nature of the writing system where only one symbol is used to cover both in loans from Greek (e.g. Greek kōthōn > Etruscan qutun "pitcher").

Voiced stops missing
The Etruscan consonant system primarily distinguished between aspirated and non-aspirated stops. There were no voiced stops and loanwords with them were typically devoiced, e.g. Greek thriambos was borrowed by Etruscan, becoming triumpus and triumphus in Latin.

Syllabic theory
Based on standard spellings by Etruscan scribes of words without vowels or with unlikely consonant clusters (e.g. cl 'of this (gen.)' and lautn 'freeman'), it is likely that were sometimes syllabic sonorants. Thus cl and lautn.

Rix postulates several syllabic consonants, namely and palatal  as well as a labiovelar spirant  and some scholars such as Mauro Cristofani also view the aspirates as palatal rather than aspirated but these views are not shared by most Etruscologists. Rix supports his theories by means of variant spellings such as amφare/amφiare, larθal/larθial, aranθ/aranθiia.

Word formation
Etruscan was inflected, varying the endings of nouns, pronouns and verbs. It also had adjectives, adverbs and conjunctions, which were uninflected.

Nouns
Etruscan substantives had five cases, a singular and a plural. Not all five cases are attested for every word. Nouns merge the nominative and accusative; pronouns do not generally. Gender appears in personal names (masculine and feminine) and in pronouns (animate, or either masculine and feminine, and inanimate or neuter); otherwise, it is not marked.

Unlike the Indo-European languages, Etruscan noun endings were more agglutinative, with sometimes two or three endings added instead of alternative endings; for example, where Latin would have distinct nominative plural and dative plural endings, Etruscan would suffix the case ending to a plural marker: Latin nominative singular fili-us, "son", plural fili-i, dative plural fili-is, but Etruscan clan, clen-ar and clen-ar-aśi.

Pallottino calls this phenomenon "morphological redetermination", which he defines as "the typical tendency ... to redetermine the syntactical function of the form by the superposition of suffices." His example is Uni-al-i, "in the sanctuary of Juno", where -al is a genitive ending and -i a locative. Steinbauer uses the term "inflecting language" (rather than inflected), which he explains as a language in which "... there can be more than one marker ... to design a case, and ... the same marker can occur for more than one case."

Nominative/Accusative Case: No distinction is made between nominative and accusative of nouns. Common nouns use the unmarked root. Names of males may end in -e: Hercle (Hercules), Achle (Achilles), Tite (Titus); of females, in -i, -a or -u: Uni (Juno), Menrva (Minerva), Zipu. Names of gods may end in -s: Fufluns, Tins; or they may be the unmarked stem ending in a vowel or consonant: Aplu (Apollo), Paa (Bacchus), Turan.

Genitive case: Pallottino defines two declensions based on whether the genitive ends in -s/-ś or -l. In the -s group are most noun stems ending in a vowel or a consonant: fler/fler-ś, ramtha/ramtha-ś. In the second are names of females ending in i and names of males that end s, th or n: ati/ati-al, Laris/Laris-al, Arn/Arn-al. After l or r -us instead of -s appears: Vel/Vel-us. Otherwise a vowel might be placed before the ending: Arn-al instead of Arn-l.

There is a patronymic ending: -sa or -isa, "son of", but the ordinary genitive might serve that purpose. In the genitive case morphological redetermination becomes elaborate. Given two male names, Vel and Avle, Vel Avleś means "Vel son of Avle." This expression in the genitive become Vel-uś Avles-la. Pallottino's example of a three-suffix form is Arnth-al-iśa-la.

Dative case: The dative ending is -si:Tita/Tita-si.

Locative case: The locative ending is -i: Tarna/Tarna-l-i.

Pronouns
Personal pronouns refer to persons; demonstrative point out: English this, that.

Personal
The first person personal pronoun has a nominative mi ("I") and an accusative mini ("me"). The second person has a dative singular une ("to thee"), an accusative singular un ("thee") and an accusative plural unu ("you"). The third person has a personal form an ("he" or "she") and an inanimate in ("it").

Demonstrative
The demonstratives are ca and ta used without distinction. The nominative/accusative singular forms are: ica, eca, ca, ita, ta; the plural: cei, tei. There is a genitive singular: cla, tla, cal and plural clal. The accusative singular: can, cen, cn, ecn, etan, tn; plural cnl. Locative singular: calti, ceii, cl(i), ecli; plural caiti, ceii.

Adjectives
Though uninflected, adjectives fall into a number of types formed from nouns with a suffix:
 * quality, -u, -iu or -c: ais/ais-iu, "god/divine"; zamai/zami-c, "gold/golden."
 * possession or reference, -na, -ne, -ni: paa/paa-na, "Bacchus, Bacchic"; laut/laut-ni, "family/familiar" (in the sense of servant)
 * collective, -cva, -chva, -cve, -ve, -ia: sren/sren-cva: "figure/figured"; etera/etera-ia, "slave/servile"

Adverbs
Adverbs are unmarked: etnam, "again"; ui, "now"; uni, "at first." Most Indo-European adverbs are formed from the oblique cases, which become unproductive and descend to fixed forms. Cases such as the ablative are therefore called "adverbial." If there is any such system in Etruscan it is not obvious from the relatively few surviving adverbs.

Verbs
Verbs had an indicative mood and an imperative mood. Tenses were present and past. The past tense had an Active voice and a Passive voice.

Present active
Etruscan uses a verbal root with a zero suffix or -a without distinction to number or person: ar, ar-a, "he, she, we, you, they make."

Past or preterite active
The -ce or -ke suffix to the root produces a third person singular active, which has been called variously a "past", a "preterite" or an "aorist." In contrast to Indo-European, this form is not marked for aspect, nor are the roots, apparently, distinguished for their aspect; they are simply actions that went on in the past. Examples: tur/tur-ce, "gives/gave"; sval/sval-ce, "lives/lived."

Past passive
The third person past passive is formed with -che: mena/mena-ce/mena-che, "offers/offered/was offered."

Vocabulary

 * See the list of Etruscan words and list of words of Etruscan origin at Wiktionary, the free dictionary and Wikipedia's sibling project

The Etruscan vocabulary is now a few hundred words known with some certainty. The exact count depends on whether the different forms and the expressions are included. The Wiktionary list referenced above is in alphabetic order. Below is a table of some of the words grouped by topic.

Some words with corresponding Latin or other Indo-European forms are likely loanwords to or from Etruscan. For example, neftś "nephew", is probably from Latin (Latin nepōs, nepōtis; German Neffe, Old Norse nefi). A few dozen from Etruscan survive in Latin; for example, elementum (letter), litterae (writing), cera (wax) (κηρός in Ancient Greek), arena (sand).

At least one word has an apparent Semitic origin: talitha "girl" (Aramaic; could have been transmitted by Phoenicians).

The Etruscan numerals are known although debate lingers about which numeral means "four" and which "six" (hu or śa). Numerals are listed in their own article. Of them, and of the basic words in general, Bonfante (1990) says: "What these numerals show, beyond any shadow of a doubt, is the non-Indo-European nature of the Etruscan language. Basic words like numbers and names of relationships are often similar in the Indo-European languages, for they derive from the same root."

General

 * Etruscan News Online, the Newsletter of the American Section of the Institute for Etruscan and Italic Studies.
 * Etruscan News back issues, Center for Ancient Studies at New York University.
 * Etruscology at Its Best, the website of Dr. Dieter H. Steinbauer, in English. Covers origins, vocabulary, grammar and place names.
 * Viteliu: The Languages of Ancient Italy at web.archive.org.
 * The Etruscan Language, the linguistlist.org site. Links to many other Etruscan language sites.

Inscriptions

 * ETP: Etruscan Texts Project A searchable database of Etruscan texts.
 * Etruscan Inscriptions in the Royal Ontario Museum, article by Rex Wallace displayed at the umass.edu site.
 * Etruscan; The Pyrgi Bilingual, paper by Michael Weiss, Cornell University.

Lexical items

 * An Etruscan Vocabulary at web.archive.org. A short, one-page glossary with numerals as well.
 * Etruscan Vocabulary, a vocabulary organized by topic at etruskisch.de, in English.
 * Etruscan-English Dictionary at iolairweb.co.uk. An extensive lexicon compiled from other lexicon sites. Links to the major Etruscan glossaries on the Internet are included.

Fonts

 * Etruscan and Early Italic Fonts, download site by James F. Patterson at webspace.utexas.edu.

Llingua etrusca Etruskeg Етруски език Etrusc Etruština Lingua etrusca Etruskisk (sprog) Etruskische Sprache Ετρουσκική γλώσσα Idioma etrusco Etruska lingvo Étrusque Lingua etrusca 에트루리아어 hsb:Etrušćina Etruščanski jezik Lingua etrusc Etrúska Lingua etrusca ეტრუსკული ენა Etruskek Etruskų kalba Etruszk nyelv Etruskisch エトルリア語 Etruskisk Etruskisk språk Język etruski Língua etrusca Этрусский язык Etruscan language Etrurski jezik Etruskin kieli Etruskiska Етруська мова 伊特魯里亞語