Pagina principală

   Listă alfabetică articole
   Ultimele articole
   Clasament articole
   Hartă articole
   Asoc. Cult. Enciclopedia Dacica (2)
   Echipa ENDA (4)
   Voluntariat ENDA (5)
   Comunicate ENDA (3)
   Rapoarte de activitate (2)
   Prezentare (1)
   Activități (8)
   Poveștile Legio Dacica (15)
   Terra Dacorum (24)
   Economia (12)
   Arta (6)
   Regii (15)
   Personaje (6)
   Societate (3)
   Origini (2)
   Triburi (82)
   Împăraţii traco-daci (1)
   Ştiinţă (1)
   Kogaionon (7)
   Armele (32)
   Seniorii războiului (7)
   Arhitectura militară (4)
   Cetăţile (20)
   Războaiele dacilor (17)
   Civis Romanus (7)
   Romanii (3)
   Dinastii imperiale (2)
   Migraţiile (10)
   Etnografica (6)
   Apoulon (5)
   Surse elene (103)
   Surse latine (140)
   Surse româneşti (97)
   Surse diverse (9)
   Lapidarium (7)
   Traduceri (177)
   Articole online
   Cărți online
   Periodice (58)
   Recenzii (11)
   Repertorii arheologice (3)
   Surse contemporane (8)
   Surse contemporane (5)
   Surse vechi (7)
   Lucrări de licenţă (2)
   Cursuri (4)
   Dacia 3D (9)
   Arheologie experimentală (3)
   Trupe de reconstituire istorică (1)
   Reconstituiri istorice (1)
   Filme artistice (4)
   Grafică (3)
   Poezii (12)
   Legende şi povestiri (3)
   English papers (55)
   Borangic Cătălin (33)
   Marcu Marius (5)
   Velico Dacus (57)
   Video (34)
   Resurse WWW (2)
   Ştiri (401)
   Diverse (2)

Pagina principalăHartă siteArhivă ştiriListă alfabetică articoleClasament articoleContact ENDA pe FacebookCanal Youtube ENDAENDA pe TwitterNoutăţi ENDA prin canal RSSAbonare newsletter Distribuie pe FacebookDistribuie pe TwitterDistribuie prin email


Alexandru Berzovan

I. Introduction

The etymology of the Dacian term dava, appearing in historical and epigraphical sources also as deva, daba etc, has preoccupied over the last century a large number of specialists. The first in-depth analysis of this term1 belongs to the reputed Austrian geographer, linguist and historian of Czech origin, Wilhelm Tomaschek (Vilém Tomášek)2. He was the author of a massive study, Die alten Thraker. Eine ethnologische Untersuchung published between 1893 and 1894 which has remained more or less a landmark in the field of Thracology (Янакиева 2016, p. 13). Thus for the Dacian term dava, Tomaschek proposed a derivation from a reconstructed Proto-Indo-European root, with some phonetic (and semantic) evolutions. A non-Indo European, Kartvelian solution has also been briefly mentioned, but dismissed as a random occurrence (Tomaschek 1893, p. 139).

More than 120 years have passed since then, and knowledge in the field of archeology, epigraphy, ancient history and comparative linguistics has grown exponentially. Nowadays, dozens of etymological dictionaries for a significant number of living or dead languages as well as a consistent amount of primary sources are accessible, permitting an extensive documentation. In the light of the above mentioned facts, we believe a critical re-evaluation of the existing hypothesis is more than necessary, a fact that we proposed to do in the current paper.

We want to say from the onset that, given the limited data we have about the languages of the local population(s) that lived in the Carpatho-Danubian area at the turn of the eras, we might not be able to come at the end with a definitive and un-objectionable solution. However, we hope that our study will prove useful by bringing into discussion some new light and necessary clarifications, while also paving way for new research directions that had not received too much attention.

II. General data
II.1. Attestations

The term -dava (with variants -daba, -deva, -deba, -dova) appears as suffix in around 50 attested placenames in Northern Thracian area. In Southern Thrace we find other suffixes for placenames such as -bara, -para (around 36-39), with -bria (around 12), -diza (around 11), while in the southern Scythia Minor we find the suffix -dina/-deina (around 12). The numbers are, of course, approximate, due to uncertainties, transliteration errors in historical sources, etc (Katičiči 1976, p. 147 -148; Janakieva 1990, p. 120-124; Olteanu 2006).

The vast majority of the placenames ending with -dava are located to the north of the Danube, in the area inhabited by the historical Dacians. They are present also on the southern banks of the Danube, a fact due most likely due to the presence of Dacians/Getae or related population in this area (see Janakieva 2014, p. 34, Karte 1), in certain particular cases probably also due to colonization of ”Transdanubian” populations by the Romans at the turn of the eras (Avram 2015, p. 146). Isolated toponyms with -dava are mentioned by Ptolemy in Sarmatia Europaea (III,5) and in Germania (II.10). Isolated is also Thermidava, attested in nowadays Albania, probably the result of a Dacian infiltration or colonization into Illyrian lands (Georgiev 1961, p. 27). The variant -deva seems to appear mostly on the southern banks of the Danube, as well as the variant -daba/doba, a fact that could be relevant for the existence of dialectal variants (Olteanu 2006).

The term does not seem to be restricted to the field of toponimy. It appears, albeit very rarely, in a number of feminine names belonging to Dacians and Thracians. We have attested a Duccidavae (compare with placename Docidava from Ptolemy, III, 8), Caridava (Dana 2014, p. 114), Ροιμηδαβα (Dana 2014, p. 114, compare with placename Ramidava from Ptolemy, III, 8), likely also in the name Davappier (Dana, Matei-Popescu p. 225; Dana 2014, p. 113-114) with the formant -pier, also specific to some Thracian and Dacian names. Naming children (especially daughters) after settlements was not something out of the ordinary among the Romanized Dacians (or any other ancient populations) see for example the case of the mother of Emperor Galerius, Romula, a refugee from Dacia, probably named after Romula in Dacia Malvensis.

Possibly according to some authors (Dana 2014, p. 114), but somewhat less probably in our opinion, is the presence of the term in personal names such as Δεβαβενζις, which seem to contain also the name of goddess Bendis (Dana 2014, p. 114), or in divine epithets such as Θεός Δαβατοπιη[νός] or Δαβατοπειός, attested in the north-eastern part of Lower Moesia (Dana 2014, p. 113), situations that could have other explanations. In the case of Δεβαβενζις we could theoretically deal with a homonymy with the local term for ”deity”, possibly related or derived from Proto-Indo-European *deywós.

Regarding the exact pronunciation, the situation is as it follows. As Olteanu pointed out (2006), the Greek transcriptions always use ε and never η, thus we are most likely dealing with short vowels, with the accent falling on the first syllable. Attempts had been made to interpret the a/e oscillation visible in -dava/-deva as the result of the existence a vowel that the Greeks and Romans found hard to pronounce and write in their scripts (Poghirc 1961, p. 36); others attempted to reconstruct an evolution from an early form -dawa towards a late form -diva (Olteanu 2006). We shall will limit ourselves to a number of observations: a) most of the literary sources and a part of the inscriptions were produced by non-native speakers of the language(s), and in most cases we do not know their sources of information and neither how precisely they perceived and rendered the sounds of the local language(s); b) in the case of the inscriptions left by natives in the Roman Empire in Latin and Greek (see above) we found mostly -dava, -δαβα or -δεβα, a fact that cannot be without significance; c) we also have to take into account the existence of various local dialects which could also account for a part the variation, a fact sustained - as we have mentioned above - by spatial distribution of -dava variants (mostly) in the north of Danube.

II.2. Chronology

Chronologically the earliest attestation of the term dava that we know so far is in the famous decree of Dionysopolis (IGB I2 13), dated sometime around the year 48 BC, honoring Akornion who traveled in a diplomatic mission to meet the father of someone (most likely of Burebista) at a location read variously as Αργεδα(υ)ον, or Ζαργεδα(υ)ον (see Vulpe 1994, p. 91 with bibliography). The earliest mention in literary sources is to be found in the Ad Urbe Condita of Titus Livius (44,26,7) ”circa Desudabam in Maedica exercitus Gallorum consederat”. This work was written in the last quarter of the 1st century BC, but the event that is being talked about happened much earlier, during the campaigns of the Third Macedonian War, in 168 BC (Janakieva 1990, p. 120-124).

Later on, the term appears in a large number of literary sources, such as the Geography of Ptolemaios, the Tabula Peutingeriana, the Lexicon of Heschyus of Alexandria, De Aedificiis of Procopius of Caesareea, and others including the Novaelle of Justinian until the late 6th - early 7th century AD. It is also consistently present in various inscriptions during the 2nd - 4th century AD period. We can thus say that we are dealing with a term attested and used on a large scale for more than half a millenia in the northern Balkan area (Chiriac 1997, p. 117-118), proving its vitality as well as its productivity, even in the context of the local identities and languages gradual dissipation in the face of increasing linguistic and cultural Hellenisation and Romanisation.

The various native languages and ethnic identities of the Northern Balkans mostly vanished by the middle of the 7th century AD (Russu 1976; Russu 1981, p. 82-83; see also Yanakieva 2018) due to the massive social, demographic, religious and ethnic reconfigurations, and the term apparently fell out of use with them. It was not preserved in the substratum of the later Balkanic languages such as Albanian and Romanian, somewhat surprising considering the Dacian or Illyrian (?) origin of the first and the vast pre-Roman substratum of the second3. Only a few isolated toponyms survived until nowadays, such as Plovdiv and Kokodiva (in Bulgaria), possibly borrowed by Slavic and Bulgar newcomers from the last speakers of Thracian or Thraco-Romans. For toponyms in Romania, the existing opinions in the specialised literature differ. For the city name Deva a derivation from -dava/deva was proposed (see Poruciuc 1995, p. 275-283; Paliga 2007b, p. 217-218) but a Slavic origin or at least a Slavic intermediary remains also as a possibility (Olteanu 2006). A Dacian origin for Turda or Arcuda, with the loss of intervocaliv -v- in Daco-Romanian was also brought into discussion by some authors (Olteanu 2006 with reference to earlier bibliography)4. The problem is complex, with various divergent opinions, sometimes with various hints of subjectivity, and we shall insist no further since the issue falls outside the scope of this study.

II.3. Meaning

A particular problem is the precise meaning of the term. Some authors considered that the term would mean ”village”, ”open settlement” (Tomaschek 1893, p. 139; Pârvan 1926, p. 72; Russu 1967, p. 29, 101), ”settlement” (Kazarow 1916, p. 29), while a few others thought that, on the contrary, it would mean ”a fortification” or a large ”fortified settlement” (Crişan 1977, p. 45)5 or an important residential center (Ursachi 1969, p. 115; Vulpe 2014, p. 297). There were also attempts to explain the term as implying the existence of a sanctuary (Nistorescu 2012, p. 39-40).

What do the ancient sources say? In the Lexicon of Heschyus (Λ 480), we have λέβα defined as Πόλις; the form λέβα is considered to be a translation error of δέβα (Velkova 1986, p. 59-60), a fact confirmed by the Lexicon of Pseudo Zonaras, were we have δαβα (Dana 2014, p. 114)6. Ptolemy, on the other hand when he discusses Dacia considers all placenames he presents (either with -dava or not) as ”polis”; but for Ptolemy even Iazygian and Sarmatian settlements are also called ”polis” (!), thus we believe that most likely we are dealing with ancient authors formalism (see Forţiu 2012, p. 27-28 with bibliography).

A look at archaeological and epigraphical evidences could bring us closer to a resolution. We have to admit - and we have all the reasons to do so - that at least a part of the davae mentioned by Ptolemy in Pre-Roman Dacia represented settlements of great importance (centers of power) that did have fortifications. For certain there were also other places without the -dava suffix in their names that also represented centers of great power and prestige (for example, the capital of ancient Dacia, Sarmizegetusa, Apoulon, and others). Furthermore, if we look further than Preroman Dacia, we find the example of Buteridava (ISM 1, 1983, p. 359-360), a vicus on the territory of Histria (Avram 1994, p. 230) that most likely did not had any fortification and was likely a rural settlement of minor importance.

All of the above strongly suggest, as already stated by some authors, that the term dava did not define specifically a ”village or open settlements”, nor did it define a ”fortification”, or a ”major settlement”. The most likely and plausible explanation is that it was a plain, generic term for ”settlement”, a fact acknowledged nowadays by most Romanian archaeologists (Glodariu 1983, p. 72-73; Sîrbu 1996, p. 111-112; see the discussion in Florea 2006, p. 248; Florea 2011, p. 169-177).

III. The Indo-European solution(s)
III.1. Derivation from Proto-Indo-European

W. Tomaschek proposed a derivation from a Proto Indo-European (PIE) root *dhê, *dhe, with the meaning ”to set”, ”to place”, expanded with a determinative ”u” (Tomaschek 1893, p. 139). Paul Kretschmer (1896, p. 222) derived from the same PIE root *dhē- with sufix ”va”, related to another root *dhā-, and thus cognate (among others) with Latin facio and Sanskrit dhitā. The same hypothesis is also found in the repertoire of D. Detschew (1957, p. 121-122); among others; Detschew proposes a further connection with a Lykian term hlmmidewe/Elmidauai that he (most likely erroneusly) interprets as a placename (Detschev 1957, p. 122)7.

The above mentioned solution has proved to be extremely popular, being taken up by a large number of authors (among them Георгиев 1957, p. 56; Duridanov 1969, p. 30; Russu 1967, p. 161; Paliga 1987, p. 24; Poghirc 1987; Paliga 2007a, p. 14; Paliga 2007c, p. 312; Мирчева, Михайлова 2016, p. 253) and it was included as such in the lexicon of reconstructed PIE roots (more recently, Starostin, Lubotksy 2007, p. 736). However, despite its wide acceptance, it presents a number of questions and problems that cannot be easily ignored.

The first problem is related to semantics. The reconstructed root *dhē-, composed of two sounds, is considered to be, apparently, at the base of a quite large number of words scattered in various Indo-European languages, both living and defunct. The meaning of the derived words varies, most of them related to the notions ”to set”, ”to place” (an object), ”to do”, ”to give”, ”seat”, ”to dress” etc. None of the words mentioned can be considered by all standards as a true ”cognate” (accordance in shape and meaning) to the Dacian dava. The second problem is that the derivation implies more than a semantic change and an extra ”formant u8; in the words of the editors of the last edition, ”Thrac. -dava ”settling, settlement” from *dhēu̯ü or *dhǝu̯ü; probably reshuffling after the concurrent *-: *dou- “bestow, give” (Starostin, Lubotksy 2007, p. 736).

In comparison, the Indo-European origin of other terms attested in Thracia for ”settlement” that we mentioned in our presentation is much more straightforward. The term -diza is cognate with Proto-Iranian *daiz- ”to build”, ”to form”, thus with Old Persian dida, ”fortress”, and Greek τεῖχος, ”a wall”, all from PIE *dheiĝh, ”to knead clay”, ”to build” (Starostin, Lubotksy 2007, p. 655-656)9. Similarly, the -dina formant is obviously cognate, if not borrowed from Celtic -dunum10.

It is not in our purpose to contest the possibility of deriving the term dava from the PIE radical *dhē-, but as we have shown this hypothesis appears to be a complicated one and marked by many assumptions that cast at least a few shadows on its plausibility. For this reason, we consider it necessary to (at least) take into account other solutions, which we will do in the following.

But before analyzing the alternative ”non Indo-European” solutions, we shall stop at discussing an opinion recently expressed by Dan Alexe (2015), who attempt to connect the Dacian term with terms from various Iranian languages. While not expressed in an academic study, and in an unsystematic manner, his opinions are enjoying at this moment a certain popularity among some historians and archaeologists of the Late Iron Age in Romania, thus we consider necessary to discuss and critically asses them.

III.2. Cognates in Iranian Languages?

Discussing and trying to explain to the larger public the problem of the Dacian language in a book that collects a number of blog entries on various subjects, he mentions that Dacian dava represents ”the Indo-European name for the wall”, thus cognate with Avestan daeza ”which is dewar in modern Persian” (Alexe 2015, p. 74). Another cognate according to him would be the Zazaki dewi, ”village” (Alexe 2015, p. 74)11.

Later on in his work, Alexe makes a short analysys about the transmission of the term in the work of Ptolemy, concluding that ”those davae, in fact, are just an indication that the cartographer's informants named like this all settlements, but not that the locals in the distant lands named them so” (Alexe 2015, p. 102). Trough this conclusion it is quite obvious that the author unfortunately did not study in depth the problem of the attestations and transmission of the term12.

Let us take a closer look at the above mentioned terms in Iranian language. The Avestan verb daez has the meaning, ”to build”, (Hashemi, Najari 2015, p. 62-63). It is cognate with Sogdian dys, ”to build”, Parthian ”dys” with the same meaning, ultimately deriving from Proto - Iranian *daiz, ”to build, to form” (see Table 1), ultimately derived in its turn from PIE *dheiĝh, ”to knead clay”, ”to build” (Cheung 2007, p. 53). The Old Persian dida ”city wall, fortress” and the Middle Persian (Pahlavi) diz, dyz, as well as New Persian (Farsi) diz ”fortress” are connected to the same root. These can be cognates with Dacian dava only if we accept that the latter is derived from the same PIE root (see our note 9).

The Middle Persian (Pahlavi) dywr, dewar, and the New Persian (Farsi) diwar, derive apparently from an unattested but supposed Old Persian formation *daida-vara (Mumm 2009, p. 24-25); its existence is almost sure, considering the Avestan vara ”fortification”, ”enclosure” (Hashemi, Najari 2015, p. 62-63; Koshelenko, Gaibov 2014, p. 87-89, etc) and dida (see above). From the Middle Persian (Pahlavi) and Modern Persian (Farsi) the term diwar spread among the Medieval Islamic ”Persianate” cultures of South-Western Asia in languages such as Urdu were we have ديوار dīwār (Platts 1884, p. 559), in Turkic Languages etc. In any case however the modern Persian diwar cannot be considered a cognate of the ancient Dacian dava, as it evolved inside the Iranian languages from older Iranian words and it is attested only since Middle Persian (Pahlavi).

Proto - IranianOld Persian (Achaemenid Period)AvestanSogdianParthianMiddle Persian (Pahlavi)New Persian (Farsi)
*daiz - to build, formdidā - city wall, fortresspairi daeza - to build, to build a fence arounddys - to builddys - to builddiz, dyz - fortressdiz - fortress
-*daida-varaVara - enclosure, fortification--dywr, dewar - fortress wall, walldiwar - village, wall

Table 1. Words for fortress, wall and village in various Ancient, Medieval and Modern Iranian languages. Forms with * are reconstruction. Table made with data taken from Cheung 2007, Mumm 2009, Hashemi, Najari 2015.

What about the Zazaki dew, can it be considered a valid cognate of Dacian dava? To answer to this problem, we have to look again for the origin of the term in Iranic Languages. Apparently, the term mentioned by Alexe exists only in modern dialects of Zazaki (Paul 1998, p. 176), being ultimately derived from Old Persian dahyu13, ”a land”, ”country”, via an intermediate form *dayw (Paul 1998, p. 176). Regarding dahyu, its meaning in Old Persian is ”land”, ”satrapy”14, the same with the Avestan daxiiu, ”land”, ”territory”, probably related with the Old Indian dasyu, ”enemy”. Its origin might lay in the proto Indo-Iranian term *das which designated a larger community of people. This root explains ethnonyms such as Dahae. It can also be compared with Khotanese term daha, meaning ”human”, ”man”. Later in Middle Persian dahyu evolves into deh ”land”, ”country”, ”village” (Gnoli 1993, p. 590), spreading in other languages such as Urdu were we have ده dih, děh (Platts, 1884 p. 537) and various South-West Asian idioms. In the light of the above mentioned consideration, neither the modern Zazaki dew nor its ”ancestor”, the Old Persian dahyu seem to be a valid cognate for Dacian dava.

We cannot exclude that in future some suplimentary data will appear that might change the interpretation we have presented here. However, we believe that such a scenario is quite unlikely, as Iranian languages, in all their stages of evolution are very well studied and well attested.

IV. The Non-Indo-European Kartvelic solution
IV.1. The Kartvelic root *dab(a)

As we mentioned in the introduction of this study, the Kartvelic hypothesis was first mentioned by W. Tomaschek, who mentiones the Georgian word daba, ”village”, ”settlement”. He puts forward (quite hastily) a root -da meaning in his opinion ”low”, ”low lying” from the word dabali, and rejected the connection with Dacian dava as a random occurence (Tomaschek 1893 p. 139). As we shall see further, his reconstruction is erroneous, based on a superficial knowledge of Kartvelian languages, understandable however considering the stage of the research at the time (late 19th century).

The Georgian term დაბა (daba) is a native Kartvelian term for ”village”, ”town”, ”settlement”; the other term, ქალაქი (kalaki), ”city”, while also attested in Old Georgian (აბულაძე 1973, p. 451) is ultimately borrowed from Armenian or another Middle Eastern idiom. The term ”daba” for village, settlements is attested in Old Georgian Bible (Xanmeti version, between 5th - 7th century AD): ”romelsa kalaksa sexwidet, anu dabnebsa” (Matthew, 10,11), ”xuxtres kalaksa da dabnebsa” (Mark, 5, 14) etc. (Fähnrich 2007, 122). It is found nowadays in a number of Georgian toponyms, sometimes as a suffix, surprisingly similar to dava/daba in Ancient Dacia (see Table 2).

Name and translationLocation(s)
დაბა Daba (”Town”)Borjomi Municipality, Samtshke-Javakheti Region
ახალდაბა Akhaldaba (”New Town”)Akhalgori Municipality, Mtskheta - Mtianeti Region
Akhmeta Municipality, Kakheti Region
Gori Municipality, Shida Kartli Region
Mtskheta Municipality, Mtskheta - Mtianeti Region
Ochamchire Municipality, Adzibubja Region
Tkibuli Municipality, Imereti Region
Shuakeva Municipality, Adjaria Region
ყველდაბა Qveldaba (”Cheese Town”)Akhalgori Municipality, Mtskheta - Mtianeti Region
მერისდაბა Meris Daba (”Major Town”)Historical Monastery (see Gamkrelidze et alii 2013, p. 331)

Table 2. Toponyms with -daba in Georgia (historical and modern placenames, non-exhaustive research)

But we do not have extent Old Georgian texts earlier than the 5th century AD when Christian literature started to be translated in the Kingdom of Kartli (see the discussion in Rayfield 1994, p. 19-27). In the Pre-Christian period the inscriptions discovered so far in the various Kartvelian kingdoms and polities were in Aramaic and Greek languages. Thus, in order to discuss any possible relation with the Dacian dava, we have to look for a much earlier timeframe than the 5th century AD.

Let us look at the other Kartvelian languages. In Mingrelian we have the term dobera, ”cornfield”, and in Svanic the term dab-, ”field” (Klimov 1998, p. 36; Fähnrich 2007, p. 122). Thus, based on this words in the existing Kartvelian languages, a Proto Kartvelian (PK) root *dab(a)- was reconstructed by Kartvelologists, with the meaning ”village”, ”cornfield” (Klimov 1998, p. 36; ფენრიხი, სარჯელაძე 2000, p. 162-163; Fähnrich 2007, p. 122). According to linguistic reconstruction, reconstructed Proto-Kartvelian (PK) was spoken before 2600 BC (Klimov 1998, IX). The semantic relation ”cornfield” / ”village” is present in numerous language families of the world, ancient and modern alike, and perfectly explainable in a society of sedentary agriculturalists like that of the Proto-Kartvelians.

A derivation of the Dacian dava from this PK root (or another archaic root that generated both the Dacian and PK one) would be simple and straightforward, without semantic or major phonetic changes, even further sustained by the existence of the variant daba for dava. As such it has already been proposed but without much discussions by a few authors (Furneé 1979 p. 31; Gordeziani 2016, p. 154).

However, the linguistic and historical problems raised by this solution are, at least at first sight, quite significant, as it would imply some sort of connections existing between the Paleo - Balcanic language(s) of the Dacians, generally considered of Indo-European stock, and the non-Indo European Kartvelian idioms spoken nowadays in North-Eastern Anatolia and the Caucasus. Are we dealing with an ”accidental convergence”, or maybe with an untraceable wanderwört? Are there any possible scenarios connecting the Southern Caucasus and the Carpatho-Danubian area? These are but a few problems that we will attempt to discuss in the following.

IV.2. A few (unlikely) scenarios of transmission

Ancient sources mention Thracians and Euboeans wandering into the Caucasus (Strabo, 11,5-2), while other sources mention Colchians at the Lower Danube, including Ovid. In Ovid´s poetry (Tristia 2, 191-192) Colchians are mentioned alongside the Iazygi and the Getae near the Danube Delta, a fact explained by some authors as the result of a later copying error (Gostar 1961, p. 314-315; Hind 2007, p. 241-245). We do not believe that we are dealing with a copyist’s error, but with the mythopoetic imagination of the exiled poet fascinated by the legend of the Argonauts; as other authors said, poetic texts should not be analyzed by employing the criteria used for historical sources, as their goal is aesthetic (Curcă 2014, p. 376). At any rate, all sources mentioning Colchians at the Lower Danube are related to the story of the Argonauts, they are legends and mythopoesis in our opinion with little if any objective historical value for our problem of interest.

Another hypothetical solution would imply the transfer of the word via Cimmerian or Scythian migrations which - as shown by archaeology - implied consistent movements of populations from the steppes, including those situated to the north of the Caucasus, into the Carpatho-Danubian area. It is generally accepted nowadays that both the above mentioned nomadic populations where speakers of Indo-European, Iranic idioms17, but this view of ethnic uniformity has been challenged in the last years. The presence of non-Iranian and non-IE terms of North-East Caucasic origin in the remaining list of Scythian words has already been pointed out (Kullanda 2014, p. 84-85). In 2014 an article written by A. Mayor, J. Colarusso and D. Saunders (2014, p. 447-493) attempted to prove that a part of the so-called ”non-sense” inscriptions on Greek vessels depicting Scythians and Amazons were in fact words explainable through various North-Pontic idioms, especially in the non-Indo-European North-Western Caucasic languages, with a few possibly Kartvelian (Mayor et alii 2014, p. 478). Despite the initial warm welcome18, the paper was subsequently criticized for various methodological flaws (Chiarini 2018, p. 199 - 200), with an especially harsh response from Russian linguist Alexei Kassian (2016, p. 181), who criticized the authors for comparing abruptly their ”readings” with terms from modern languages and not with the reconstructed forms of North-West-Caucasian proto-languages which would have been much closer to what was spoken in the area 2500 years ago. At any rate, for our problem of interest, the ”Scythian / Cimmerian” solution has too many flaws. One is that the Kartvelian people were not present in large numbers in the North-Caucasian area, nor did they engage in seasonal or permanent nomadism, like their northern neighbors. The terminology of reconstructed Proto-Kartvelian does not suggest this. While it is plausible that among the Iranic Scythians or the Agatyrhsoi or the earlier Cimmerians could have been some speakers of non-Indo European languages of the Caucasus, it is unlikely they were Kartvelian, and even more unlikely they had the strength to influence the language(s) of the sedentary populations that they encountered in the Carpathians, even less the term related to the notion of settlement.

IV.3. An incursion in the problem of the pre-Indo-European substratum

The last hypothesis we will discuss that could (theoretically) explain a relation between the Dacian dava and the PK *dab, is the most plausible and it implies an incursion into the deep prehistory. The problem of pre-Indo-European languages spoken in the Balkan Peninsula, Anatolia and their relations with the languages of the Caucasus has over time concerned a large number of specialists. The links between various ”Mediterranean” languages, the pre-Greek substrate and the languages of the Caucasus, with a special focus on Kartvelian, have been discussed since the early 20th century, from many perspective, in a number of works of varying scientific quality19.

An important landmark were the studies of Dutch linguist and Classicist Edzard Furnée (1972), who analyzed the pre-Indo-European elements of the Greek language, and as we mentioned earlier, included the term dava in his lists of Pre-Greek - Kartvelian correspondences (Furneé 1979 p. 31). This specific direction of investigation by comparing elements of the pre-Greek with Kartvelian enjoyed and apparently still enjoys a certain popularity in Georgian historiography, especially in the studies of Rismag Gordeziani who was studying the problem since the late 1960s (Gordeziani 1969, p. 11-21)20. Attempts have been made to explain various obscure names of Greeks deities and heroes or toponyms from the Pre-Greek substratum by comparing them with Kartvelian, with more or less convincing results (a few recent examples: Gordeziani 2006, p. 105-116; Darchia 2006, p. 44-52; Chotalishvili 2007, p. 53-63; Chotalishvili 2014, p. 63-72).

It is true that up to this moment, in the context when comparison between various languages of the Caucasus and other distant language families have almost become mainstream, these points of view received little comment or support (Tuite 2008b, 49, note 34). In our opinion, this situation might be due to several few reasons. The languages of the Caucasus are generally little known in the field of Classical studies21, and at least in the case of Kartvelian there might be a certain restraint to discuss theories that were subject to strong nationalistic ideologisation in Georgia in the early 1990s during the controversial regime of Zviad Gamsakhurdia22. And generally speaking, there are few specialists in the domain; in Romania at least, the languages of the Caucasus represent an exotic subject with no specialists to our knowledge and no available literature23.

Even if some aspects of the theories presented by Gordeziani and his students might be debatable, at least a part of his material, like that of Furneé is of great value, a fact accepted by other specialists (for ex. Gamkrelidze, Ivanov 1995, p. 799-804). It is almost certain that in one way or another the Proto-Kartvelian, or to use Furneé´s terminology, rather Paleo-Kartvelian (Furneé 1979, p. 17) populations communed with other pre-Indo-European or ”Mediterranean” people (Rayfield 2012 p. 11-12).

Returning to our specific problem of interest, could there be Proto-Kartvelian parallels in Dacian language, other than the presumed dava? Are there linguistic arguments to suppose the influence or presence of Proto-Kartvelian(like) elements in the Northern Balkans? Unfortunately for most of the remaining terms of Dacian language, a large majority toponyms and personal names (thus, of debatable relevance) we do not know their exact meaning, and comparing them abruptly with Proto-Kartvelian roots (or with any other language, in our opinion) would likely lead to numerous false-positive results and false conclusions, in short, a methodological disaster. A more prudent possibility to respond to the above mentioned question - although not without its shortcomings - would be to see if the substratum elements of Romanian presents Kartvelian parallels25. When one knows the semantics of the elements that are compared, the relevance of the comparison gains a completely different weight. Furthermore, there is every reason to presume that most of the Romanian substratum elements are of local, Paleo-Balkanic origin, even if some might not be specifically ”Dacian”.

We have to say that comparisons of Romanian substratum elements with Kartvelian had already been made (see Lahovary 1963, p. 182, p. 209 and others; more recently Mosenkis 2016, p. 26-28; Leschber 2016, p. 243, p. 245), although some of the chosen examples are debatable26. A closer examination we made at the Romanian language substratum - words that are defined as ”substratum”, and words with unknown etymology that could not be explained in mainstream etymological dictionaries as recent borrowings, offered a number of interesting parallels with Proto-Kartvelian roots and Kartvelian languages, converging both semantically and formally27. Even if a few of these correspondences might prove at a closer analysis to be accidental in nature, at least some of these words could be accepted as pre-Indo-European, ”Mediterranean” in origin, if not specifically ”Kartvelian-like”. A few of them seem peculiar only to Romanian, Albanian and Kartvelian or only to Romanian and Kartvelian. A more thorough discussion would go beyond the scope of this study.

Yet we might ask ourselves how did these words enter into the substratum of the Romanian language. Where they present as substratum in the Indo-European tongue of the Dacians or other related tribes?29 Brought in by Roman colonists? Or, more likely, adopted from the language of smaller, obscure Carpathian tribes, ”Dacian” in material culture but still speaking peculiar archaic tongues30, that underwent Latinization in the period after 106 AD? It is difficult to answer at this moment.

Returning to our problem, how did the contacts between Proto or Paleo-Kartvelian and Balkans idioms, like ancient Greek, happened? If T. Gamkrelidze and V. Ivanov attempted to explain these correspondences in the light of the myth of the Argonauts (?), generally in the framework of the ”Armenian theory” regarding to the origin of Indo-European people (Gamkrelidze, Ivanov 1995, p. 799-804)31, other authors have proposed different scenarios.

Rismag Gordeziani proposed a migration of the Proto-Kartvelian tribes in the 3rd - 2nd millenniums BC from nowadays Western Georgia, towards eastern Europe on the northern shores of the Black Sea and another migration into the Aegean area (Gordeziani 2011, p. 49, 52; Gordeziani 2015, p. 101-102)32. Unfortunately, up to now, there are no reliable historical sources, or even less archaeological ones to account for such a migration of people, even if individual or indirect contacts are plausible. The possibility of ties between the Early Iron Age Colchian culture and Early Iron Age cultures from the Carpatho-Danubian space, based on object analogy and similarity in ceramic style had been brought into attention by archaeologist Joni Apakidze (2006, p. 19), but these hypothesis, while theoretically possible, requires a much more solid archaeological argumentation (Sagona 2018, p. 451).

The problem is ultimately related to the ethnogenesis of the Kartvelian people, a subject that remains quite obscure (Baramidze 2017, p. 323-332), marked by various diverging opinions. The possibility of a reverse migration, of proto-Kartvelian tribes from the central and western Anatolia towards Transcaucasia has been brought into discussion as a possibility by certain authors (Kavtaradze 1997, p. 352-357; Kavtaradze 2001, p. 43-44), with the possibility of Proto-Kartvelians forming from the mixture of ”Old European” people33 displaced from the Carpatho-Danubian area and local Anatolian tribes at the beginning of the Bronze Age (see Kavtaradze 2000, p. 107, 115)34.

V. Final considerations

At the end of this study, we are left with the question: what is, in fact, the etymology of the Dacian term dava? As we anticipated in the introduction, we do not have a definite answer to this problem. The Indo-European solution, while it presents its shortcomings, does not raise major historical problems and does not challenge the existing paradigms about ancient Thrace and Ancient Dacia. We are convinced that for a long time to come it will continue to be cited and accepted as such as it represents the ”orthodox” solution.

The ”Kartvelian solution”, even if it’s more straightforward - and in our personal opinion in the light of the existing data cannot be any more easily dismissed or ignored- obliges to a change of paradigms. It may imply a much more nuanced view on the linguistic make-up of the Carpatho-Danubian area in prehistory, that could not be anymore reduced to the accepted evolutive scheme of Indo-Europeanisation - Proto-Thracians - Thracians - Dacians still considered classical, at least in Romanian historiography35. It may also imply the existence of a non-Indo-European component which would consist of more than a few obscure toponymical residues, strong enough to influence the developing Dacian language, if not to coexist side by side with it. It would not be unprecedented or unexpected, because in other areas of Ancient Europe, where ample historical and ethnographical sources do exist, we find such survivals up until the Roman Period (Iberian Peninsula, Southern Gaul, The Alps, Central Italy, Sicily, Sardinia, the Aegean and so forth). It remains for further studies to complete these observations.

Figure 1. Areal distribution of settlements with suffixes -dava, -deva, -daba, etc, after Janakieva 2014. Note: the present map is orientative; the real location of a large part of these settlements, especially those located north of Danube, remains largely controversial.

1 For a short list of older etymologies, of which many might appear quite fanciful today, see Tocilescu 1880, p. 240.
2 For a few biographic details, see Killy, Vierhaus 2006, p. 65.
3 The problem of the Pre-Roman substratum of Romanian represent a delicate problem, that unfortunately receives less and less attention in the last decades. Estimates of the number of substratum words vary between a few hundreds and a few thousand words. There are also ”maximalist” views, often with some taints of subjectivism bordering ”Dacomania” (for example see Vinereanu 2008), that generally received harsh criticism (Boerescu 2011, p. 291-293). We share the opinion expressed by the late Sorin Olteanu, who estimated about 1000 or more substratum words in the Romanian language (Olteanu 2006, note 19). Besides his training in linguistics, Olteanu, who was a member of the ”Vasile Pârvan” Institute of Archaeology in Bucharest, had a deep knowledge of ancient history, classical languages and epigraphy. Unfortunately, his untimely death prevented him from publishing much of his materials.
4 In general, the ancient Dacian and Roman city names have not been preserved north of the Danube. This is most likely due to the harsh political, military and economic situation that befell the native Dacian-Roman population after the retreat of the Roman administration (devastations caused by the Hunnic invasion, by the Plague of Justinian, the extreme climate events of the 6th century, Slavic and Avar migrations etc.) that ultimately reduced it during the 7th - 8th centuries AD to a number of enclaves mostly in the mountainous and forested area, especially in the Western Carpathians, with the open and the lower areas dominated by various migrating people (Madgearu 2019, p. 13-16 with the bibliography).
5 Opinions criticized by John Nandriş (1980, p. 124).
6 And thus, the assumption expressed by some researchers that λέβα would represent another Thracian name for ”town” (Paliga 2007a, p. 15-16) has to be rejected.
7 In more recent literature, this Lykian term is implied to denote a grave-protecting deity (Neumann 2007, p. 95), thus the final compound dewe is more likely to derive from PIE *deywós.
8 A fact considered problematic by more than one researcher (see the discussion in Poruciuc 1995; Мирчева, Михайлова 2016, p. 253).
9 More recently, a derivation from a Proto Indo-European *dhēgwā was proposed for the term dava by K. Witczak (2006, p. 493; 2012, p. 163), having as cognate the name of Thebes, with an (somewhat atypical) evolution of PIE *g(h)w to v in Thracian. It is not very clear if the author refers to the PIE root *dhegwh, ”to burn”, ”day” (Starostin, Lubotsky 2007, p. 652) or to the root *dheiĝh ”to knead clay”, ”to build” (Starostin, Lubotsky 2007, p. 655) or (more likely) another root that he reconstructed independently. The comparison with the name of Thebes is problematic, since is verly likely that the name of this famous Greek city is not Indo-European, probably Semitic or from other non-Indo-European pre-Greek idiom (see Poruciuc 1995 with the bibliography).
10 And maybe not coincidentally in Scythia Minor we do find Celtic placenames, such as Noviodunum, possibly Durostorum, etc (see Falileyev 2007, p. 1-28). The problem is, of course, delicate, since archaeologically speaking, there are few if any pre-Roman Celtic traces in this region.
11 According to our bibliography, dewi means ”villager”, the Zazaki term for village being dew (Paul 1998, p. 176).
12 We appreciate Dan Alexe as a writer and we believe he has plenty of talent in literary fiction, satire, as well as a solid culture; we also strongly sustain his efforts in combating pseudo-history and pseudo-archaeology. However, he lacks methodology when approaching more serious subjects, using mostly a journalistic method to obtain his answers (at least in what he published until now). Heavily influenced in his writings by 19th century Russian and French travel literature, for D. Alexe historical linguistics remains mainly as a tool to further his political essays and literary fictions.
13 As Tomaschek also correctly observed (Tomaschek 1893 p. 139).
14 A similar etymology, from dahyu and ab (Iranian word for water) was proposed for the placename Dauaba, mentioned by Ptolemy and most likely located somewhere on the banks of the Caspian Sea (see RE, IV, 2, 1901, p. 2230).
15 Xanmeti/Chanmeti is one of the dialects specific to Old Georgian Language specific to the earliest manuscripts, (see Tuite 2008a, p. 145-146).
16 For a discussion and definition of the notion wanderwörten in historical linguistics, see Haynie et alii 2014, p. 2-4.
17 For Cimmerians at least, a generic ”Caucasic” or even Kartvelian origin has been proposed in older scientific literature, but these hypotheses are no longer considered actual (see the discussion in Lebedynsky 2004, p. 9-10, 159).
18 It was featured in the National Geographic, see (accessed on 23.02.2020).
19 Starting from the works of N. J. Marr, which claimed to have detected Caucasic or, as he called them ”Japhetic” elements in various ancient and modern languages of Southern Europe; as Kevin Tuite observed, the proposed historical scenario was not in itself absurd, but was severely undermined by exclusive reliance on wildly implausible onomastic etymologies, and after 1923, by a bizarre and unscientific ”theory of language” (see the discussion in Tuite 2008b, p. 41-44; Tuite 2011). In Romania the theses of Marr had an echo in the works of linguist and Indo-Europenist Gheorghe Ivănescu (1957, p. 199-231; Ivănescu 1968, p. 273-276) but without the eccentric overtones (see the discussion in Poruciuc 2014, p. 314-316). Probably one of the most complex attempt to distinguish a common ”Mediterranean” substratum by comparing various non-Indo-European languages from Dravidian India to the Pyrenees belongs to Nicholas Lahovary (see Lahovary 1963), a member of the Romanian exile. Although his book received at the time of its apparition criticism for numerous factual errors, generalizations, and some abrupt interpretations (for ex. Zgusta, Zvelebil 1961, p. 127-130; Hester 1968, p. 223-225 and others), Lahovary gathered a lot of valuable material that is still occasionally quoted in scientific literature dealing with the Paleo Balkan substratum (more recently, Leschber 2015, p. 32-37). Unfortunately, in Romania his works were mostly ignored (see discussion in Poruciuc 2014, p. 314, note 6). The problem of possible prehistoric relations between the Balkan and South Caucasian area were also briefly discussed by A. Poruciuc (1992, p. 9-8). In more recent years, attempts had been made to define a ”Circumpontic Substratum” visible in Balkan languages, Turkic, Persian, Georgian and others, without a conclusive argumentation (Loria-Rivel 1998a, p. 47-49; Loria-Rivel 1998b, p. 50-51; Loria-Rivel 2000, p. 50-51; Loria-Rivel 2000, p. 13-17; Loria-Rivel 2002, p. 237-240; Loria-Rivel 2015, p. 44-51).
20 Unfortunately, despite our insistences, we could not access most of the articles and books published by Rismag Gordeziani before 2000.
21 In the latest etymological dictionary of Ancient Greek, when discussing the problem of the Pre-Greek substratum, Robert Beekes (2010) states that ”The comparison with Basque or Caucasian languages has not been considered in this dictionary, as this is not my competence; it is likely that there are such connections, but this must be left to other scholars.” (p. XV).
22 Who despite not having a training in history or linguistics, supported the ”Mediterranean - Kartvelian” hypothesis in its most exuberant forms (Tuite 2008b, p. 64-66), creating an ideological atmosphere surprisingly similar to the nationalistic ”Dacianism” that existed in Romania during the regime of Nicolae Ceauşescu (for the ”Dacianism”, see Mihăilescu - Bîrliba 1997, p. 157-160; Niculescu 2005, passim; Vulpe 2014, p. 294-295).
23 In other eastern European countries (for example, Poland) there are journals and centers of study dedicated to Georgian language and culture. Paradoxically, in a country whose literary and religious history was so profoundly marked by the activity of Antim the Ivirian (ანთიმოზ ივერიელი), there is not a single reliable bilingual dictionary nor a textbook to be used by those interested in learning the Georgian language.
24 In fact, some authors have already proposed Kartvelian etymologies for the name of the river Istros (Gordeziani 2011, p. 50-51), and for the name of the Carpathians (Mosenkiis 2016, p. 22, in a work that otherwise combines sometimes very interesting observations with much speculation). Since we do not know the exact meaning of these terms in the native language(s), we can neither accept or reject the hypothesis. It is interesting to note, however, the presence of Istros/Istria river names in Crete (Gordeziani 2011, p. 50-51), but also in Crimean Peninsula, possibly also in Kuban (Peţan 2011, passim) a distribution which would indicate to an archaic origin (whatever that would be).
25 The major problem is that Romanian become a literary language at a relative late date in history and the ancient native ancient language(s) of the Carpatho-Danubian area are poorly attested, thus reconstructing the ancient form of the substratum elements via sound laws is almost impossible. The problem becomes even more complex if we take into account that if some languages (and terms) suffered ample phonetic changes over time, others have remained surprisingly stable; due to lack of data we cannot say precisely in which category the Romanian substratum elements fit. In any ways, operating with words attested in modern language and comparing them with ancient forms or reconstructed forms in various proto-languages, even if widely practiced in numerous scientific studies, might not be the soundest solution in terms of methodology and it presents numerours risks (see for example in our text the discussion about the so-called Iranian parallels for dava); in cases such as this, is, unfortunately, unavoidable.
26 For example, comparing the Romanian word măgură (”a hillock”, ”a hummock”) with modern Georgian magali (Lahovary 1963, p. 209; Leschber 2016, p. 243) is debatable, since the latest could be a borrowing from Byzantine Greek since is not present in the etymological dictionaries of Kartvelian languages that we consulted (that focus only the inherited vocabulary). Any way, according to our personal, very limited knowledge of the Georgian language, much closer to magali, ”large”, great” would be the Romanian (substratum?) măgăoaie with identical meaning.
27 We do not want to propose etymologies, we mention these parallels as a possibility; this list is far from exhaustive, we stop at mentioning just a few of the more obvious. Abbreviations used: CK = Common Kartvelian, GZ = Georgian - Zan: acira, acera ”to wait, to hope, to persevere”, compare with CK *čer-, *čir-, ”to hold, to cath, attain, reach” (Klimov 1998, p. 320); agorî (rare, regional), ”to shake a tree”, with Mingrelian verb go-raua, ”to shake of the nut trees” (Klimov 1998, p. 157), maybe accidental convergence since according to Klimov root is CK *req, or more recent borrowing (?); argea ”sunken dwelling”, with Svanic argi, ”home” (Wardrop 1911, p. 610) and Svanic argil ”a place” (Gudjedjiani, Palmaitis 1985, p. 64), analogies in other languages are semantically problematic (Boerescu 2018, p. 15); abur, ”steam”, ”fog”, with CK *bur, ”to cover”, ”to darken” (Klimov 1998, p. 20) and CK *ber, ”to blow”, ”to inflate” (Klimov 1998, p. 11); bleancă ”a slap” compare with GZ *blikw, ”fist” (Klimov 1998, p. 17); brăbui, ”to bewitch”, ”to spell”, compare with GZ *bir, ”to entice, lure”, see also Georgian bireba, and Mingrelian birapa (Klimov 1998, p. 15); brâglă, ”piece from the weaving warp, with which the thread inserted with the shuttle is beaten” with GZ *breg, ”to strike”, ”to knock” (Klimov 1998, p. 19); cioară, ”crow”, compare with CK *kwer, especially Svanic čwer, with identical meaning (Klimov 1998, p. 93), very likely from the “Mediterranean” substratum since similar terms are also attested in various Italian dialects and other areas (Ungureanu 2017, p. 97; Boerescu 2018, p. 37-38); cheie ”a gorge, ravine”, compare with GZ *qew ”ravine, chasm”, see also Georgian khevi with the meaning ”gorge” (Klimov 1998, p. 335), and pre-Roman (likely pre-IE) Sardinian keia (Wagner 1997, p. 269-270); daşu, ”lamb”, Arom. gaciu, ”ram” (Popescu -Sireteanu 2017, p. 79) with CK *wac, especially Svanic ɣwaš, ”mountain goat, capricorn” (Klimov 1998, p. 50); dârlău, ”a dangler”, probably from a root dârlă, or dâr (?) compare with CK *dar-,dr-”to be unfit”, ”bad” (Klimov 1998, p. 37); drui, ”a piece of wood”, ”log”, compare with CK *dwire, also Svanic dir, with the same meaning (Klimov 1998, p. 40) also has parallels in Indo-European, Slavic, Greek (Derksen 2008, p. 122-123); hău, ”a ravine, chasm, abyss”, compare with GZ *qew ”ravine, chasm” (Klimov 1998, p. 335); gâgău, a type of shepherds food, with CK *gab, ”to cook, to boil” (Klimov 1998, p. 25); gurguiat (Aromanian), ”round”, compare with *CK grgw, ”round artefact, ring”, derivation with gurg for ”round” are specific to other languages of the Caucasus also (Klimov 1998, p. 32); moş ”an old man”, with PK *meč, ”old man”, ”old age” (Fähnrich 2007, p. 288); mugur, ”bud”, with GZ *kukul with same meaning (Klimov 1998, p. 104); muşat, Aromanian mşat, ”beautiful”, with CK *m-šwen with identic meaning (Klimov 1998, p. 128-129); pârşă, obscure word with meaning ”bad meat”, ”peeled skin”, with PK *prc-wn, ”to peel”, *prec, ”to tear”, ”rend” (Klimov 1998, p. 203-204); tihăraie, tîhlă, obscure Romanian words suggesting a root *tihă with the meaning ”thick and dense forest”, compare with CK (?) *tqe, ”forest”, ”wood” (Klimov 1998, p. 193); ţâr, ţâră, ”small”, ”little” with GZ *mcire, with same meaning (Klimov 1998, p. 129); ţâclă, ”too for cathing fish” with CK *txewl, ”to catch, to look for”, especially Georgian txevl, ”to catch fish” (Klimov 1998, p. 78); ţurlă, ”unmarried woman”, with CK *zura, ”woman”, Svanic zural, with paralels in other languages of the Caucasus (Klimov 1998, p. 61); ţurţur, ”icycle”, ”gutter through which water flows” with GZ *çurçul, ”to pearl, murmur, to stream” (Klimov 1998, p. 316-317) and so forth. Most of Romanian words mentioned are rare regionalisms, of a residual charachter in the modern language, main source for them was the work of Dorin Ştef (2015) and It is interesting to note that Slavic and Latin also seem to present some sporadic convergences with Proto-Kartvelian (Gamkrelidze, Ivanov 1995, p. 776 note 18; Klimov 1998, passim).
28 Even if between the 15th century until the 19th century Romanian transhumant shepherds did seasonally visit the Caucasus including the territory of nowadays Georgia, being present as far as Batumi and Tbilisi (Dragomir 1938, p. 159-160), direct linguistic borrowings from modern Kartvelian languages into Romanian are less likely.
29 Apparently, this is the line of interpretation of the authors that discussed possible pre-Indo-European elements in the Romanian substratum (see Paliga 2006, passim).
30 The unity of material culture visible archaeologically does not necessarily mean a single language, the problem being much more complicated, a fact amply debated in modern archaeological theory (see for ex. Chrisomalis, Trigger 2003, p. 419-433). Against such a scenario could be brought the testimonies of ancient sources, which tell us only about the Getae and Dacians as local, ”indigenous” populations in the Carpathian area. However, we must keep in mind that the ancient sources that discuss ”barbarian” lands ussualy pay attention only to the groups that were representative from a political and military point of view; furterhmore, the sources we have about Dacia are generally of questionable quality - unfortunately, the best historical sources such as the works of Emperor Trajan, Criton, Dyo Chrisostomos and others which very likely contained many valuable historic, ethnographic and linguistic data have been lost.
31 For a criticism of the theories regarding a so-called Anatolian origin of the Indo-European languages, see Haarman 2007, p. 159-162.
32 Similar ideas also in the later works of E. Furnée (1986, especially), which we could not consult, unfortunately.
33 Themselves the result of earlier successive migrations from Anatolia and Southern Balkans in the Neolithic period.
34 Without commenting too much on these hypotheses, we limit ourselves to saying in passing that the Proto-Kartvelian languages were in contact for a long time with the Proto-Indo-European idioms (see Smitherman 2012 with bibliography). A hypothetic presence of Proto-Kartvelian(like) elements in the Carpatho-Danubian space in the late Eneolithic were they could have interacted with their eastern Proto-Indo-European neighbors would give no more reasons to deny the ”Kurgan Theory”, nowadays contested almost exclusively on the grounds of the linguistic fact we mentioned in the first phrase.
35 For a recent, critical assessment of the Romanian historiography dedicated to the problem of Indo-Europeanisation, with valuable observations, see Gogâltan 2016.


We thank Professor Dr. Adrian Poruciuc (Iaşi Institute of Archaeology / Alexandru Ioan Cuza University of Iaşi), who helped us with suggestions and bibliography throughout the process of writing this study.
Alexandru Berzovan


აბულაძე 1973 - ი. აბულაძე, ძველი ქართული ენას ლექსიკონი (Old Georgian Lexicon), მეცნიერება, თბილისი, 1973.
Alexe 2015 - D. Alexe, Dacopatia şi alte rătăciri româneşti, Ed. Humanitas, Bucureşti, 2015.
Apakidze 2006 - J. Apakidze, Westlische Kontakte der Kolchis-Kultur im 2. und Anfang des 1. Jahrtausends v.Chr., in Phasis, 9, 2006, p. 14-27.
Avram 1994 - A. Avram, Buteridava, in C. Preda (Ed.), Enciclopedia Arheologiei şi Istoriei Vechi a României, I, Ed. Enciclopedică, Bucureşti 1994, p. 230.
Avram 2015 - A. Avram, Aspects de la colonisation des Daces au sud du Danube par les Romains, in I Traci tra geografia e storia, Trento, 2015, p. 143-160.
Baramidze 2017 - T. Baramidze, ”Dark Ages” in Comparative Studies and Ethnogenesis of Georgians in the Framework of the History of Comparative Studies, in Linguistics and Literature Studies, 5,5, 2017, p. 323-332.
Beekes 2010 - R. Beekes, Etymological Dictionary of Greek, Brill, Leiden-Boston 2010.
Boerescu 2011 - P. Boerescu, Mihai Vinereanu. Dicţionar etimologic al limbii române pe baza cercetărilor de indo-europenistică, Bucureşti, 2008, 2009, 935 p, in Limba Română, LX, 2 2011, p. 291-293.
Boerescu 2017 - P. Boerescu, Etimologii româneşti controversate. Ipoteze şi soluţii, Ed. Academiei, Bucureşti 2017.
Boerescu 2018 - P. Boerescu, Elementele de substrat (autohtone) ale limbii române. Compendiu, Ed. Academiei, Bucureşti, 2018.
Cheung 2007 - J. Cheung, Etymological Dictionary of the Iranian Verb, Brill, Leiden-Boston 2007.
Chiarini 2018 - S. Chiarini, The So-called Nonsense Inscriptions on Ancient Greek Vases, Brill, Leiden / Boston 2018.
Chiriac 1997 - C. Chiriac, Unele observaţii asupra informaţiilor literar-istorice bizantine privitoare la regiunea Dunării de Jos în secolele V-X, in Arheologia Moldovei, 20, 1997, p. 107-126.
Chotalishvili 2007 - L. Chotalishvili, The Colchian Heritage of Helios against the Background of Linguistic Data, in Phasis, 10, 1, 2007, p. 53-63.
Chotalishvili 2014 - L. Chotalishvili, On Cretan - South Caucasian Toponymic Parallels, in Phasis, 17, 2014, p. 64-72.
Chrisomalis, Trigger 2003 - S. Chrisomalis, B. Trigger, Reconstructing Prehistoric Ethnicity: Problems and Possibilities, in J. Wright, J. Pilon (eds), A Passion for the Past, Papers in Honor of James F. Pendergast, Archaeological Survey of Canada, Mercury Series Paper no 164, Canadian Museum of Civilization, Gartineau, 2003, p. 419-433.
Crişan 1977 - I. H. Crişan, Burebista şi epoca sa, Ed. Ştiinţifică şi Enciclopedică, Bucureşti, 1977.
Curcă 2014 - R. Curcă, Greek and Latin Authors on the Carpathian - Dniestrian Territory: An Anthropology of Perceptions, in Studia Antiqua et Archaeologica, XX, 2014, p. 369-380.
Dana 2014 - D. Dana, Onomasticon Thracicum, Editions de Boccard, Athenes, 2014.
Dana, Matei-Popescu 2009 - D. Dana, F. Matei-Popescu, Soldats d'origine dace dans les diplomes militaires, in Chiron, 39, 2009, p. 209-256.
Darchia 2006 - I. Darchia, Some Lexical Peculiarities of Crete`s Ancient Population Dialect, in Phasis, 9, 2006, p. 44-52.
Detschew 1957 - D. Detschew, Die thrakischen Sprachreste, Wien, 1957.
Derksen 2008 - R. Derksen, Etymological Dictionary of the Slavic Inherited Lexicon, Brill, Leiden, 2008.
Dragomir 1938 - N. Dragomir, Oierii mărgineni în Basarabia, Caucaz şi America de Nord, in Lucrările Institutului de Geografie al Universităţii din Cluj, 6, 1938, p. 159-303.
Duridanov 1969 - I. Duridanov, Thrakisch - Dakische Studien, I, Verlag der Bulgarischen Akademie der Wissenschaften, Sofia, 1969.
ფენრიხი, სარჯელაძე 2000 - ჰ. ფენრიხი, ზ. სარჯელაძე, ქართველურ ენათა ეტიმოლოგიური ლექსიკონი, თბილისი, 2000.
Fähnrich 2007 - H. Fähnrich, Kartwelisches Etymologisches Wörterbuch, Brill, Leiden, 2007.
Falileyev 2007 - A. Falileyev, Celtic Dacia. Personal names, place-names, and ethnic names of Celtic origin in Dacia and Scythia Minor, Aberystwyth, 2007.
Florea 2006 - G. Florea, Oppidum, Dava. Câteva observaţii terminologice, in C. Gaiu, C. Găzdac, Fontes Historiae. Studia in honorem Demetrii Protase, Ed. Accent, Cluj-Napoca, 2006, p. 245-249.
Florea 2011 - G. Florea, Dava et Oppidum. Débuts de la genèse urbaine en Europeau deuxième âge du Fer, Académie Roumaine, Centre d’Études Transylvaines, Cluj-Napoca, 2011.
Forţiu 2012 - S. Forţiu, Ziridava în context ptolemeic, Ed. Sfântul Ierarh Nicolae, Brăila, 2012.
Furnée 1972 - E. Furnée, Die wichtigsten konsonantischen Erscheinungen des Vorgriechischen, De Gruyter Mouton, 1972.
Furnée 1979 - E. Furnée, Vorgriechisch - Kartvelisches, Editions Peeters, Leuven, 1979.
Furnée 1986 - E. Furnée, Paläokartvelisch-Pelasgische Einflüsse in den Indogermanischen Sprachen, Hakuchi Press, Leiden, 1986.
Георгиев 1957 - В. Георгиев, Тракииският Език, Издание на Българската Академия на Науките, София, 1957.
Gamkrelidze, Ivanov 1995 - T. Gamkrelidze, V. Ivanov, Indo-European and the Indo-Europeans, De Gruyter, Berlin-New York 1995.
Gamkrelidze et alii 2013 - G. Gamkrelidze, D. Mindorashvili, Z. Bragvadze, M. Kvachadze, A topoarchaeological dictionary of Kartlis Tskhovreba (The Life of Kartli / History of Georgia), Georgian National Museum, Tbilisi, 2013.
Georgiev 1961 - V. Georgiev, Sur L`Ethnogenese des Peuples Balkaniques, Le Dace, L`Albanais et le Roumain, in Studii Clasice, 3, 1961, p. 23-27.
Glodariu 1983 - I. Glodariu, Arhitectura dacilor, Ed. Dacia, Cluj-Napoca, 1983.
Gnoli 1993 - G. Gnoli, Dahyu, in Enciclopedia Iranica, VI, 6, London, 1993.
Gogâltan 2016 - Fl. Gogâltan, Transilvania şi problema Indo-Europenizării. Paradigma românească, in Annales Universitatis Apulensis, Historica, 20, 2, 2016, p. 11-42.
Gordeziani 1969 - R. Gordeziani, Zur Frage der Agaisch-Kartwelischen Sprachparallelen, in Wissenschaftliche Zeitschrift der Friedrich-Schiller-Universitat, 18, 5, 1969, p. 11-21.
Gordeziani 2006 - R. Gordeziani, The Cult of Dionysus in the Light of Linguistic Data, in Phasis, 9, 2006, p. 105-116.
Gordeziani 2011 - R. Gordezianu, The Colchians in the Adriatic - Poetic Imagination or a Historical Fact, in Phasis, 13-14, 2010-2011, p. 42-54.
Gordeziani 2015 - R. Gordeziani, Pre-Greek-Caucasian Hypothesis at the Modern Stage of Research, in Phasis, 18, 2015, p. 100 -110.
Gordeziani 2016 - R. Gordeziani, Pre-Greek-Etruscan-Caucasian Language Relations, lecture sustained at ”The Ancient World and the Caucasus” Winter School, Ivane Javakhishvili Tbilisi State University, Tbilisi, 2016, (accesed on 23.02.2020).
Gostar 1961 - N. Gostar, ”Metereaque turba” (Ovide, Tristia, II, 191), in Studii Clasice, 3, 1961, p. 314-315.
Gudjedjiani, Palmaitis 1985 - Ch. Gudjedjiani, L. Palmaitis, Svan - English Dictionary, Caravan Books, New York, 1985.
Haarman 2007 - H. Haarman, Indo-Europenization - the seven dimension in the study of a never-ending process, in Documenta Praehistorica, XXXIV, 2007, p. 155-175.
Haynie et alii 2014 - H. Haynie, C. Bowern, P. Epps, J. Hill, P. McConvell, Wanderwörter in languages of the Americas and Australia, Ampersand, 1, 2014, p. 1-18.
Hashemi, Najari 2015 - M. Hashemi, H. Najari, Avestan Architecture: A Descriptive Etymological Lexicon, in International Letters of Social and Humanistic Sciences, 63, 2015, p. 59-70.
Hester 1968 - D. A. Hester, Recent Developments in Mediterranean ”Substrate” studies, in Minos, 9, 1968, p. 219-235.
Hind 2007 - J. Hind, Ovid and the Barbarians beyond the Lower Danube (Tristia 2. 191-2; Strabo, Geogr. 7.3.17), in Dacia, N.S., LI, 2007, p. 241-245.
ISM 1 - D. M. Pippidi, Inscripţiile din Scythia Minor greceşti şi latine. 1. Histria şi împrejurimile, Ed. Academiei, Bucureşti, 1983.
Ivănescu 1957 - Gh. Ivănescu, Le rôle des Japhétites dans la formation des peuples et des cultures antiques, in Studia et Acta Orientalia, I, 1957, p. 199-231.
Ivănescu 1968 - Gh. Ivănescu, Albanais et japhétites, in Studia et Acta Orientalia, VII, 1968, p. 273-276.
Janakieva 1990 - S. Janakieva, Apparition, diffusion et variantes des noms de lieux thraces à deuxième thème -ΔΑΒΑ -ΔΑVΑ, in M. Taceva, D. Boiadijev, Studia in Honorem Borisi Gerov, Sofia Press, 1990, p. 120-124.
Janakieva 2014 - S. Janakieva, Thrakisch und Dakisch - Sprachen Oder Dialekte? in Orpheus, 21, 2014, p. 21-37.
Kavtaradze 1997 - G. Kavtaradze, The Interrelationship between the Transcaucasian and Anatolian populations by the data of the Greek and Latin literary sources, in P. Roman, S. Diamandi, M. Alexianu (Eds.), The Thracian World at the Crossroads of Civilizations, Semne 94, Bucureşti, p. 352-361.
Kavtaradze 2000 - G. Kavtaradze, Some Problems of the Interrelation of Caucasian and Anatolian Bronze Age Cultures in Quaderni di Archeologia, Universita de Messina 1,1, 2000, p. 107-124.
Kavtaradze 2001 - G. Kavtaradze, Two Trascaucasian Ethnonyms of Anatolian Origin, in Caucasian and Near - Eastern Studies, X, Ivane Javakhishvili Institute of History and Ethnology, Mematiane, Tbilisi, 2001, p. 39-56.
Kassian 2016 - A. Kassian, Un-making Sense of Alleged Abkhaz - Adyghean Inscriptions on Ancient Greek Pottery, in Ancient Civilizations from Scythia to Siberia, 22, 2016, p. 177-198.
Katičiči 1976 - R. Katičiči, Ancient Languages of the Balkans, 1, Mouton, 1976.
Kazarow 1916 - G. Kazarow, Beiträge zur kulturgeschichte der Thraker, Studnička, Sarajevo, 1916.
Koshelenko, Gaibov 2014 - G. Koshelenko, V. Gaibov, The Avestan Vara and the Early Towns of Central Asia, in Parthica, 16, 2014, p. 69-91.
Kretschmer 1896 - P. Kretschmer, Einleitung in die Geschichte der Griechischen Sprache, Göttingen, 1986.
Kullanda 2014 - S. Kullanda, External relations of Scythian, in Journal of Language Relationship, 11, 2014, p. 81-90.
Killy, Vierhaus 2006 - W. Killy, R. Vierhaus, Dictionary of German Biography (DGB), 10, K.G. Saur, München.
Lahovary 1963 - N. Lahovary, Dravidian Origins and the West, Orient Longmans, Madras, 1963.
Lebedysnky 2004 - I. Lebedynsky, Les Cimmeriens, Editions Errance, Paris, 2004.
Leschber 2015 - C. Leschber, Implications of Macro-Areal Linguistics, in Slavia Meridionalis, 15, 2015, p. 32-41.
Leschber 2016 - C. Leschber, On the stratification of substratum language, in B. Hansen, B. Whitehead, Th. Olander, B. Olsen, Etymology and the European Lexicon. Proceedings of the 14th Facthagung der Indogermanischen Gesellschaft, 17-22 September 2012, Copenhagen, Reichert Verlag, Wiesbaden, 2016, p. 235-250.
Loria-Rivel 1998a - G. Loria-Rivel, Persian in the Context of Circumpontic Languages. Presumed Vestiges of a Suffixed Definite Article, in Thraco-Dacica, 19, 1-2, 1998, p. 47-49.
Loria-Rivel 1998b - G. Loria-Rivel, Direct Object Marking in Circumpontic Languages and Turkish, in Thraco-Dacica, 19, 1-2, 1998, p. 50-51.
Loria-Rivel 2000 - G. Loria-Rivel, Bulgarian Indirect Speech Verbal Forms and the Circumpontic Substratum, in Thraco-Dacica, 21, 2000, p. 13-17.
Loria-Rivel 2002 - G. Loria-Rivel, ”Have to-“ Sentences of Compulsion in Circumpontic Languages, in Proceedings of the VIIIth Congress of Thracology, ”Thrace and the Aegean”, I, Institute of Thracology, Sofia, p. 237-240.
Loria-Rivel 2015 - G. Loria-Rivel, The Conjunctive Article in Circumpontic Languages and its Possible Pre-Indo-European Substratum, in П. Асенова, P. Станчева, B. Алексова, P. Бейлери (Eds.), Балканските езици, литератури и култури Дивергенция и конвергенция, София, 2015, p. 44-51.
Madgearu 2019 - A. Madgearu, Expansiunea maghiară în Transilvania, Ed. Cetatea de Scaun, Târgovişte, 2019.
Mayor et alii 2014 - A. Mayor, J. Colarusso, D. Saunders, Making sense of non-sense inscriptions associated with Amazons and Scythians on Athenian vases, in Hesperia, 83, 2014, p. 447-493.
Mihăilescu-Bîrliba 1997 - V. Mihăilescu-Bîrliba, Impact of political Ideas in Romanian Archaeology before 1989, in Studia Antiqua et Archaeologica, 3-4, 1997, p. 157-160.
Мирчева, Михайлова 2016 - A. Мирчева, Б. Михайлова, Етимологични бележки върху няколко тракийски глоси, in Е. Търпоманова, Б. Михайлова, В. Алексова (eds.), Балканското Езикознание Днес. Сборник в чест на 75-годишния юбилей на проф. д.ф.н. Петя Асенова, Университетско издателство ”Св. Климент Охридски”, Sofia, 2016, p. 247-257.
Mosenkis 2016 - Iu. Mosenkis, Hellenic Origin of Europe, PP Zhovtyi O. O. Edition, Kyiv / Uman 2016.
Mumm 2009 - P. Mumm, Strukturkurs Neupersisch, 2009, (accessed at 15.09.2019)
Nandriş 1980 - J. Nandriş, Crişan, I. H., Burebista and his Time, Biblioteca Historia Romaniae, Monograph XX, Editura Academiei Republicii Socialiste România, Bucharest, 1978, 253 pp, 22 lei, in The Slavonic and East European Review, 58, 1 (Jan., 1980), p. 124-126.
Neumann 2007 - G. Neumann, Glossar des Lykischen, Harrasowitz Verlag, Wiesbaden, 2007.
Niculescu 2005 - Gh. Niculescu, Archaeology, Nationalism and ”The History of the Romanians”, in Dacia, NS, XLVIII-XLIX, 2004-2005, p. 99-124.
Nistorescu 2012 - L. Nistorescu, Criterii extrafilologice în analiza etimologică. Observaţii pe marginea formantului toponomastic dava, in Philologia Banatica, IX, 2, p. 33-43.
Olteanu 2006 - S. Olteanu, Categorii de toponime în funcţie de origine şi aşezare, 2006, (accesed on 02.02.2020)
Paliga 1987 - S. Paliga, Thracian terms for ”Township” and ”Fortress”, and Related Place-Names, in World Archaeology, 19, 1, 1987, p. 23-29.
Paliga 2006 - S. Paliga, Etymological Lexicon of the Indigenous (Thracian) Elements in Romanian, Ed. Evenimentul, Bucureşti, 2006.
Paliga 2007a - S. Paliga, Thracian terms for ”township” and ”fortress” and related place names, in S. Paliga, Etymologica et Anthropologica Maiora, Ed. Fundaţia Evenimentul, Bucureşti, 2007, p. 13-20.
Paliga 2007b - S. Paliga, Ten Theses on Romanian Etymology, in Etymologica et Anthropologica Maiora, Ed. Fundaţia Evenimentul, Bucureşti, 2007, p. 209 - 232.
Paliga 2007c - S. Paliga, Despre Taba / Teba, Dava / Deva, despre alte aspecte ale fondului pre-indo-european, ale celui indo-european, ale celui proto-boreal, despre ”nostratisme” precum şi despre coerenţă în tracologia lingvistică, in Etymologica et Anthropologica Maiora, Ed. Fundaţia Evenimentul, Bucureşti, 2007 p. 299 - 314.
Paul 1998 - L. Paul, The Position of Zazaki among West Iranian Languages, in N. Sims - Williams (ed), Proceedings of the Third European Conference of Iranian Studies, I, Dr. Ludwig Reichert Verlag, Wiesbaden, 1998, p. 163-177.
Pârvan 1926 - V. Pârvan, Getica. O Protoistorie a Daciei, Cultura Naţională, Bucureşti, 1926.
Peţan 2011 - A. Peţan, Omonimele Istrului. 1. Istrul Taurilor, in A. Peţan, A. Berzovan (eds), Arheologie şi studii clasice, 1, Ed. Dacica, Cluj-Napoca, 2011, p. 235-267.
Platts 1884 - J. T. Platts, A Dictionary of Urdu, Classical Hindi, and English, W. H. Allen & Co., London,1884.
Poghirc 1961 - C. Poghirc, Le valeur phonetique de l`oscillation graphique thrace a/e a la lumiere des donees des langues balkaniques modernes, in Studii Clasice, 3, 1961, p. 33-37.
Poghirc 1987 - C. Poghirc, Noms traces et geto-daces de habitation et leur survivances balkaniques, in Cahiers Balkaniques, 12, 1987, p. 260-270.
Popescu-Sireteanu 2017 - I. Popescu - Sireteanu, Termeni păstoreşti în limba română, 1, Ed. Alfa, Iaşi, 2017.
Poruciuc 1992 - A. Poruciuc, Problems and patterns of the Southeast European Ethno- and Glottogenesis (cca. 6500 BC - AD 1500), in The Mankind Quarterly, XXXIII, 1, 1992, p. 3-42.
Poruciuc 1995 - A. Poruciuc, The etymology of Dac. deva in connection with Gk. Thebai and with Paleobalkan phonology, in Thraco-Dacica, 16, 1-2, p. 275-283.
Poruciuc 2014 - A. Poruciuc, Fenomenul indoeuropean în viziunea lui Gheorghe Ivănescu, in Anuar de lingvistică şi istorie literară, LIII, p. 313-322.
Rayfield 1994 - D. Rayfield, The Literature of Georgia. A History, 2nd edition, Routledge, 1994.
Rayfield 2012 - D. Reyfield, Edge of Empires. A History of Georgia, London, 2012.
RE 1901 - G. Wissowa (Ed.), Paulys Realencyclopädie der classischen Altertumswissenschaft, Vierter Band, Zweiter Halbband, IV,2, Stuttgart, 1901.
Russu 1967 - I. I. Russu, Limba traco-dacilor, Ed. II, Ed. Ştiinţifică, Bucureşti, 1967.
Russu 1976 - I. I. Russu, Elementele traco-getice în Imperiul Roman şi în Bizantium, Ed. Academiei, Bucureşti 1976.
Russu 1981 - I. I. Russu, Etnogeneza românilor, Ed. Ştiinţifică şi Enciclopedică, Bucureşti, 1981.
Sagona 2018 - A. Sagona, The Archaeology of the Caucasus. From Earliest Settlements to the Iron Age, Cambridge University Press, 2018.
Smitherman 2012 - T. Smitherman, On Ancient Kartvelian - Indo-European lexical contacts and their consequences for Proto - Indo - European, in B. Nielsen - Whitehead, T. Olander, B. Anette-Olsen, J. Rasmussen, The Sound of Indo-European. Phonetics, Phonemics and Morphophonemics, Museum Tusculanum Press, Copenhagen, 2012, p. 501-522.
Sîrbu 1996 - V. Sîrbu, Dava getică de la Grădiştea, jud. Brăila, Ed. Porto Franco, Brăila, 1996.
Starostin, Lubotksy 2007 - G. Starostin, A. Lubotsky (Eds.), Proto-Indo-European Etymological Dictionary. A Revised Edition of Julius Pokornky`s Indogermanisches Etymologisches Wörterbuch, Indo-European Language Association, 2007.
Ştef 2015 - D. Ştef, Dicţionar de regionalisme şi arhaisme din Maramureş, Ed. Ethnologica, Baia Mare, 2015.
Tocilescu 1880 - Gr. Tocilescu, Dacia înainte de romani, Tipografia Academiei Române, Bucureşti 1880.
Tuite 2008a - K. Tuite, Early Georgian, in R. Woodard (ed), The Ancient Languages of Asia Minor, Cambridge University Press, New York, 2008, p. 145-165.
Tuite 2008b - K. Tuite, The Rise and Fall and Revival of the Ibero-Caucasian Hypothesis, in Historiographia Linguistica, XXXV:1/2, 2008, p. 23-82.
Tuite 2011 - K. Tuite, The reception of Marr and Marrism in the Soviet Georgian Academy, in F. Mühlfried, S. Sokolovsky (eds.), Exploring the Edge of Empire: Soviet Era Anthropology in the Caucasus and Central Asia, LIT Verlag, 2011, p. 197-214.
Ungureanu 2017 - D. Ungureanu, Cuvinte de substrat? Da. Cuvinte dacice? Nu, in Journal of Humanistic and Social Studies, 8, 1, 2017, p. 91-101.
Ursachi 1969 - V. Ursachi, Contribuţii la problema aşezărilor dacice de pe Valea Siretului, in Memoria Antiquitatis, I, 1969, p. 105-119.
Vinereanu 2008 - Mihai Vinereanu, Dicţionar etimologic al limbii române pe baza cercetărilor de indo-europenistică, Ed. Alcor, Bucureşti, 2008.
Velkova 1986 - Z. Velkova, The Thracian Glosses. Contribution to the Study of the Thracian Vocabulary, Amsterdam, 1986.
Vulpe 1994 - A. Vulpe, Argedava, in C. Preda (Ed.), Enciclopedia Arheologiei şi Istoriei Vechi a României, I, Ed. Enciclopedică, Bucureşti 1994, p. 91.
Vulpe 2010 - A. Vulpe, Impactul cercetării civilizaţiei geto-dacice asupra societăţii contemporane, in Angustia, 14, 2010, p. 293-298.
Wagner 1997 - M. Wagner, La Lingua Sarda: storia, spirito e forma, Ilisso Edizioni, Nuoro, 1997.
Wardrop 1911 - O. Wardrop, English - Svanetian Vocabulary, in Journal of Royal Asiatic Society, 1911, p. 589-634.
Witczak 2006 - K. Witczak, The Phonological Curiosities of the Thracian Language, in Linguistique Balkanique, XLV, 3, 2006, p. 487-497.
Witzack 2012 - K. Witzack, Studies in Thracian vocabulary (I-VII), in Studia Indogermanica Lodziensia, VII, 2012, p. 153-168.
Yanakieva 2018 - S. Yanakieva, Language and Ethnic Identity of the Thracians during the Roman Age and Late Antiquity, in L. Vagalinski, M. Raycheva, D. Boteva, N. Sharankov, Proceedings of the First International Roman and Late Antique Thrace Conference, National Archaeological Institute with Museum, Sofia, 2018, p. 131-140.
Янакиева 2016 -C. Янакиева, 120 Години Тракииско Езикознание Етапи и Перспективи, in Journal of the Bulgarian Academy of Sciences, 2, 2016, p. 13-17.
Zgusta, Zvelebil 1961 - L. Zgusta, K. Zvelebil, Review of four works by N. Lahovary, in Archiva Orientalni, 29, 1961, p. 127-130.

Keywords: Dacian language; Proto-Indo-European; Proto-Kartvelian; Kartvelian Languages; Romanian substratum.
Abstract: The etymology of the Dacian word ”dava” has preoccupied over the last century a large number of specialists. In the first in-depth etymological analysis of the term, W. Tomaschek proposed a derivation from Proto-Indo-European. A non-Indo-European Kartvelian solution was briefly mentioned but dismissed as random occurrence. The first hypothesis had nowadays become almost mainstream in the field of linguistic Thracology. More than 120 years have passed since then, and knowledge in the field of archeology, epigraphy, ancient history and comparative linguistics has grown exponentially. Nowadays, dozens of etymological dictionaries for a significant number of living or dead languages as well as a consistent amount of primary sources are accessible, permitting an extensive documentation. In the light of the above mentioned facts, we believe a critical re-evaluation of the existing hypothesis is more than necessary, a fact that we proposed to do in the current paper. The Dacian word most likely represented a simple, generic term for ”settlement”. The derivation from Proto-Indo-European, seems complicated, marked by assumptions that cast a few shadows on its validity. It still remains plausible, but we believe we should look at other solutions as well. Regarding the Kartvelian hypothesis, in the light of more recent studies dedicated to this language group, it appears much more simple and straightforward. However, the Kartvelian solution raises a number of historical and linguistic problems as it would imply some sort of connections existing between the Paleo - Balcanic language(s) of the Dacians, generally considered of Indo-European stock, and the non-Indo-European idioms spoken nowadays in North-Eastern Anatolia and the Southern Caucasus. The problem is delicate, and has to be viewed within the larger context of the relationships existing between the Pre-Greek substratum and Kartvelian. There is also the possibility of connections between elements from the Romanian language substratum and the Kartvelian languages. A possible scenario could imply the existence of a non-Indo-European component in the Carpatho-Danubian area in Antiquity which would consist of more than a few obscure toponymical residues, strong enough to influence the developing Dacian language, if not to coexist side by side with it. It would not be unprecedented or unexpected, because in other areas of Ancient Europe, where ample historical and ethnographical sources do exist, we find such non-Indo-European survivals until the Roman Period (Iberian Peninsula, Southern Gaul, The Alps, Central Italy, Sicily, Sardinia, the Aegean and so forth).

Cuvinte cheie: limba dacilor; proto-indo-europeană; proto-kartvelică; limbi kartvelice; substratul limbii române.
Rezumat: Etimologia termenului dacic ”dava” a preocupat de-a lungul ultimului veac un număr mare de specialişti. În prima analiză etimologică amănunţită dedicată acestui termen, W. Tomaschek propune o derivare din Proto-Indo-Europeană. O soluţie non-indo-europeană, kartvelică este menţionată succint şi respinsă, fiind considerată rod al întâmplării. Prima ipoteză a fost acceptată aproape unanim în tracologia lingvistică. Au trecut mai bine de 120 de ani de atunci iar cunoştinţele în domeniul arheologiei, epigrafiei, lingvisticii comparate, au crescut exponenţial. Astăzi dispunem de zeci de dicţionare etimologice pentru un număr însemnat de limbii vii şi moarte, ca de altfel şi de posibilitatea accesării unui mare număr de surse de informare. În lumina acestor fapte considerăm necesară o reevaluare critică a acestor ipoteze, lucru pe care ne-am propus să îl facem în lucrarea de faţă. În lucrarea noastră discutăm atestările şi sensul cuvântului dacic care a reprezentat un termen generic pentru ”aşezare”, ”localitate”. Derivarea din Proto-Indo-Europeană este complicată, marcată de presupuneri ce ridică anumite semne de întrebare cu privire la validitatea ei. În ce priveşte ipoteza kartvelică, în lumina studiilor recente dedicate acestei familii de limbi, ea ne apare cu mult mai simplă şi mai directă. Cu toate acestea, aceasta soluţie ridică o serie de probleme istorice şi lingvistice însemnate, de vreme ce implică existenţa unor conexiuni între limbile paleo-balcanice vorbite de daci, considerate în general a fi parte din familia limbilor indo-europene şi limbile non-indo-europene vorbite actualmente în Anatolia nord-estică şi Caucazul de sud. Problema este una delicată şi trebuie privită în contextul discuţiilor referitoare la corespondenţele dintre substratul pre-grec şi limbile kartvelice. De asemenea, există posibilitatea unor conexiuni între elemente din substratul limbii române şi limbile kartvelice. Un posibil scenariu ar putea implica existenţa unei componente non-indo-europene în spaţiul carpato-dunărean la nivelul Antichităţii ce ar consta în mai mult de câteva obscure reziduri toponimice, suficient de puternică încât să influenţele limbii dacilor, dacă nu chiar să coexiste cu ea. Un astfel de scenariu nu ar fi deloc surprinzător, întrucât în alte spaţii ale Europei Antice, pentru care dispunem de ample surse istorice şi etnografice, găsim supravieţuiri pre-indo-europene până în plină epocă romană (Peninsula Iberica, Gallia sudică, Munţii Alpi, Italia Centrală, Sicilia, Sardinia, Marea Egee etc.)


Titlu complet: Considerations regarding the etymology of the Dacian term dava/deva/daba. A historical and linguistic journey from the Lower Danube to Anatolia and Transcaucasia
Autor: Alexandru Berzovan
Apărut în: ”Tracii şi vecinii lor în antichitate: arheologie şi istorie = The Thracians and their neighbours in antiquity: archaeology and history: studies in honor of Valeriu Sîrbu at his 70th anniversary”, ed.: Ionel Cândea, Brăila, Editura Istros a Muzeului Brăilei ”Carol I”, 2020, ISBN 978-606-654-403-0, p. 97