2.1. Anatolian

2.1.1. Anatolian as archaic Indo-European

Proto-Anatolian was the earliest language to branch off of the parent Proto-Indo-European trunk. This can be seen, for example, in the following archaic traits, compared to Late Proto-Indo-European (LPIE) innovations (Kloekhorst 2008):

·       While thematic o-stems were already productive in PIA, some stems show a previous stage, such as the reinterpretation of athematic PIA *heku- ‘horse’, reconstructed from Anatolian, into LPIE *hekos. While thematisation of an athematic stem is a simple step, the opposite is infrequent.

·       Retention of original uvular fricatives where LPIE evolves to pharyngeal fricatives. Laryngeal evolution different from LPIE (§II.2. Laryngeal evolution) although the colouring of neighbouring vowels is similar.

·       Animate (common) vs. neuter gender, in contrast with the opposition of feminine to masculine among animates in LPIE.

·       The fourth number, collective plural, is still fully productive for animate nouns in Anatolian, which shows a number of collective pluralia tantum.

·       Case system with certain potential defective (e.g. ablative and dative plural) or archaic inflections—obscured by later dialectal developments —not undergoing LPIE innovations, especially regarding the thematic and pronominal inflections (e.g. lack of distinction of singular and plural in oblique cases of personal pronouns).

·       Full development of neuter heteroclite nouns in *-r̥/-(e)n-, *-tr̥/-t(e)n-, *-mr̥/-m(e)n-, *-sr̥/-s(e)n-, *-u̯r̥/-u̯(e)n-, etc. (Rieken 1999)

·       Verbal system with certain archaic traits, such as two tenses (past vs. preterite, from a predecessor of the LPIE present/aorist opposition), two aspects (imperfective with ske-verbs and perfective), two moods (indicative and imperative), two voices, and two conjugations (with the innovative hi-conjugation), as well as the participle. Most of these are derived from an archaic PIE stage ancestral to LPIE, with some traits being innovations only found within Anatolian.

·       Hitt. mer-zi/mar- ‘to disappear’, cognate with LPIE *mer- ‘to die’, must point to the original meaning, since the semantic development of disappear as a euphemism for dying is much more likely than the reverse.

·       PA opposition *tiH, obl. *tu- seems more likely to reflect the original situation, which would have undergone in LPIE the adoption of the general oblique form for *tuH.

·       Hitt. šāḫ-i ‘to fill up, to plug, to stuff’, cognate with LPIE *seħ- ‘satiate’ is also more likely the original meaning of the verb.

·       PA points to PIA nom. *dégχt ‘daughter’, while LPIE material points to *dhugħtḗr, which is probably a derived form from the declension of the former.

·       PIE *χerɣw- can be reconstructed as meaning ‘to plough’, but Proto-Anatolian material points to an original ‘to crush (the ground)’, which suggests that Anatolian split off before the introduction of the plough.

·       PIA verbal root *meh- ‘to refuse, to reject’ is found in LPIE only as the 2sg. imp. act. form *meh ‘don’t!’, grammaticalised as a prohibitive particle.

2.1.2. Anatolian evolution and contacts

In Proto-Anatolian, the following phonetic changes can be seen:

·       The old PIE laryngeal system collapses (Kloekhorst 2008):

o   PIA *χ, *ɣw were preserved in some environments.

o   PIA *χ, *ɣw PA */H/ in positions *#He- and *CRHV. In all other positions merge of *ɣw, *h and loss.

o   Allophonic colouring of pre-PA *e due to adjacent *χ and w becomes phonemicised, yielding PA *a and *o.

·       PIA *eh PA *ǣ.

·       Monophthongisation of *ei and *eu, and of *oi, *ai, *ou, and *au in some environments.

·       Geminate nasals, liquids, and stops arise through assimilation.

·       Probable merger of the voiced aspirates with voiced stops.

·       Voicing of IE voiceless stops after long accented vowels and in unaccented syllables.

·       PIA medial *kw PA *gw except before *s.

·       Affricate *ts- < PIA *t-.

Suvorovo chiefs are probably to be identified with Proto-Anatolian speakers expanding from Khvalynsk, and were thus in close contact with the (most likely Proto- or Para-Uralic-speaking) Sredni Stog culture, and with cultures from the Caucasus and Old Europe, which makes any innovative trait traced to the Proto-Anatolian stage suspicious of being a potential loan.

Traits associated with early contacts could include the following:

·       The satemising trend proposed for Anatolian (Melchert 1987), if accepted, could stem precisely from this close contact (see below §3.4.1. Indo-Iranian evolution and §4.13.1. Balto-Slavic evolution).

·       Similarly, the ‘fortis-lenis’ system Pre-PA **tt/t/ˀt → PA *tt/t (Kloekhorst 2008) may stem from early contacts with languages of the Caucasus.

·       PU common structure noun + ending + poss. enclitic is found exclusively in Anatolian, which suggests a common origin in Indo-Uralic (Kloekhorst 2008), but possibly also its adoption by Pre-Proto-Anatolian migrants:


noun + ending + -mV + ending

noun + case suffix + -mV


noun + ending + -tV + ending

noun + case suffix + -tV


noun + ending + -sV + ending

noun + case suffix + -sV


The earliest attested Anatolian language is possibly to be found in the inscriptions of Armi, dated ca. 2500-2300 BC (Bonechi 1990), whose onomastic tradition is used to locate it in or near Ebla territory, in what is today north-western Syria (Archi 2011):

“Most of these personal names belong to a name-giving tradition different from that of Ebla; Arra-ti/tulu(m) is attested also at Dulu, a neighbouring city-state (Bonechi 1990b: 22–25). We must, therefore, deduce that Armi belonged to a marginal, partially Semitised linguistic area different from the ethno-linguistic region dominated by Ebla. Typical are masculine personal names ending in -a-du: A-la/li-wa-du/da, A-li/lu-wa-du, Ba-mi-a-du, La-wadu, Mi-mi-a-du, Mu-lu-wa-du. This reminds one of the suffix -(a)nda, -(a)ndu, very productive in the Anatolian branch of Indo-European (Laroche 1966: 329). Elements such as ali-, alali-, lawadu-, memi-, mula/i- are attested in Anatolian personal names of the Old Assyrian period (Laroche 1966: 26–27, 106, 118, 120).”

Common Anatolian seems to have expanded thus early during the 3rd millennium BC into the three known main groups, due to their close relatedness: Southern Anatolian (comprising Luwian and Lycian, and probably Lydian), and two conservative branches, Palaic and Hittite. Intensive language contact after the spread of Common Anatolian is apparent from the morphological and phonological convergence of different dialects, which makes their classification more difficult.

The first attested Hittite and Luwian words come from clay tablets unearthed at Kaneš ca. 19201720 BC, before the first texts written in Hittite. Written in Old Assyrian dialect of Akkadian, the tablets refer to the local Anatolian population, and record hundreds of personal names that may be related to various languages, including Hittite, Luwian, Hurrian, and Hattian. The merchant records contain a number of Anatolian Indo-European loanwords adopted by the Assyrian community.

Hittite loans include layers of Hattic, Hurrian, Akkadian loanwords. Potential substrates behind some Anatolian languages include (Watkins 2001):

·         Phonetic changes, like the appearance of /f/ and /v/.

·         Split ergativity: Hurrian is ergative, Hattic probably too.

·         Increasing use of enclitic pronoun and particle chains after first stressed word: in Hattic after verb, in Hurrian after nominal forms.

·         Almost obligatory use of clause initial and enclitic connectors: e.g. semantic and syntactic identity of Hattic pala/bala and Hittite nu.

Interesting is the Indo-Iranian words found in the hippological texts of Kikkuli, which contains e.g. PII. aikaartanna- ‘single turn’, maybe through Luwian or Hurrian (see below §3.4.4. Mitanni Indic). The two last layers seen on Hittite are Luwian-like (the so-called “Glossenkeilwörter”, marked by writers as of foreign origin), and the Luwian loanwords increasing in the Middle Hittite, and especially in the Neo-Hittite periods.

Luwian loans include potential Hittite Luwianism PII assussanni-, as well as Lycian esbe, assumed to derive from the Mitanni reflex of LPIE *ekos ‘horse’.


2.1.3. Schleicher’s fable in Proto-Anatolian

hwou̯is ʔku̯es-hwe

hwou̯is ku̯oisom hlʔni no ʔest

ʔkums ʔaus;

kom densom u̯ogom ugtm̥,

kom m̥kom borom,

kom hduhsom ʔāku brtm̥.

to teʔt hwou̯is ʔku̯os:

ʔāgor-mu kēr,

pesenom ʔtsi ʔkums neihnihtm̥.”

to tent hku̯es: “klu(dhi) hwu̯e!

ʔāgor-nos kēr ʔtsi,

pesōn, ʔesos, hwu̯i̯ōm hlʔni

tu̯ekəm ʔāitatəm u̯osbom hwnetsi,

hwu̯i̯ōm-hwe hlʔni no ʔestsi.”

Tod keklu̯ts hwou̯is pl̥hnom nuntri̯et.

For Proto-Anatolian vocabulary, Kloekhorst (2008) has been used.

·       No common verbs for ‘carry’, ‘bear’, or ‘listen’ are found in Anatolian, especially one that may be etymologically related to common LPIE verbs. Since we have selected LPIE *u̯egh, *bher-, and *kleu- for PIU and PIA versions, the most likely output of these roots in Proto-Anatolian have been used.

·       PIA *pleχnom has been selected for ‘field, land’ because LPIE *χegros seems to be absent from Anatolian languages, and a connection to Sumerian agar ‘irrigated territory, grainfield’ has been proposed, also found in Semitic *hagar, and maybe NE Caucasian *ˀüććürV ‘meadow, glade, clearing’.