1.1 The Palaic Texts

Palaic is the least attested member of Forrer’s group of the acht Sprachen der Boghazköi-Inschriften, after the sociolinguistically minor Mittani Aryan. Like the other ‘local,’ non-Mesopotamian members of this group, Palaic appears to be documented exclusively in the Hittite archives, which means that all we read is filtered by the scribal habits and praxis of the official scribes of Hattuša. Therefore, graphemic laws that apply to Hittite also apply to Palaic: we may assume that the rendering of consonant clusters reflects the conventions used by the Hittites; we can assume that Sturtevant’s law¹ applies and renders consonantal phonemic oppositions much as in Hittite;² we can also interpret the notations of ambiguous vowels (i/e alternation) and plene writing much as we do in Hittite (as a not always consistent way to note the vowel length, which was probably linked to accent position in a complex fashion).³ Yet, as pointed out by Kudrinski (2017), the Palaic texts are encoded somewhat differently than Hittite texts when it comes to Akkadograms and Sumerograms. The former are virtually absent and the latter very infrequent. This pattern is sociolinguistically unproblematic, as it mostly matches that attested for Luwian (in which Akkadograms are very rare and Sumerograms rare, but a bit more frequent than in Palaic), although it appears to be more extreme (but a different amount of available documents forces us to be cautious when assessing the available data), and argues that scribes who took dictation in this language had a limited competence in it and wrote the syllables they heard without replacing them with the usual heterograms.⁴

The Palaic texts found in the Hittite archives are very limited in number and seem to be exclusively connected to the cultic sphere. In recent years, Palaic fragments have been assigned CTH numbers from 751 to 754 and some new joins have been identified (mostly by D. Sasseville).⁵ Although a detailed description would risk rapid obsolescence due to the importance of the philological works currently in progress, the nature of the main texts and text groups that carry evidence of the Palaic language can be summarized. CTH 751, 752, and 753 are Festrituale for what is sometimes called the ‘Palaic pantheon,’ with the significant difference that, in the current tentative partition, CTH 751 seems to collect ritual offers, while CTH 752 and some fragments cataloged under CTH 753 appear to involve the Anatolian topos of the ‘disappearing god.’⁶ At this point is appropriate to introduce a well-known problem that will be mentioned again in this chapter: the difficulty of disambiguating Palaic and Hattian cultural and religious constructs. The divinities mentioned in the texts that we wish to ascribe to the Palaic pantheon include figures such as Ziparfa/Zaparfa, Katahzifuri, and Hašamili, who not only occur in the Hattian tradition but in some cases have Hattian names (see below § 2.3). Other divine names appear to be Anatolian (Šaušhalla, Hašauwanza, Aššanuwanta, Hilanzipa) and might originate from a Hittite or Luwian linguistic environment. A few of the names do exhibit a Palaic morphological marker (such as the suffix -ika- in the Ilaliyantikeš and GUL-/Gulzannikeš divine groups and the possible Luwian suffix -anni- in the latter), but it is hard to trace these theonyms to a specific non-Hittite and non-Luwian Palaic lexicon (in other words, they could easily represent Palaicized versions of foreign god names). CTH 754, containing further Palaic fragments, concludes the inventory of texts in Palaic. Other texts, such as CTH 750 (a ritual related to the god Ziparfa/Zaparfa), CTH 643 (fragments mentioning the god Ziparfa/Zaparfa), and some portions of larger festivals (nuntarriyašha and AN.TAH.ŠUM), are in Hittite but related to the Palaic cultic milieu.⁷

In the documents that carry direct Palaic linguistic material, the Palaic text (while occasionally framed by Hittite passages) is generally monolingual. This situation, consistent with the use of Palaic as a religious language employed in ritual and cultic contexts, makes the role of linguistic analysis crucial for the Entschlüsselung of the language. Accordingly, the scholarship on Palaic began in the context of Indo-European studies. After Forrer’s announcement that Palaic existed (based on the occurrence of the adverb palaumnili ‘in Palaic’ in CTH 750, a composition that, however, does not contain Palaic passages), more than two decades passed before the first Palaic text, KUB 35.165 (CTH 751), was properly published by Otten (1944). Otten recognized that Palaic was an Indo-European language, but it was not until the late 1950s that a grammatical study appeared (Kammenhuber 1959a and 1959b, a thesaurus and grammar, respectively). Subsequently Carruba (1970) published a short corpus and grammar, which is the most complete study of Palaic grammar available, although the philological ordering of the fragments is outdated. General Indo-Europeanists, made aware of the importance of this new language, by now recognized as belonging to the Anatolian branch of the Indo-European family, made significant contributions to its study (Watkins 1975 and Szemerényi 1979 on historical morphology; Wallace 1983 on sound changes affecting the vowel system). The historical phonological evidence was systematized in the dedicated chapters in Melchert (1994:190–228), which remains the standard reference almost 30 years later. Studies on specific texts or topics have supplemented Melchert’s diachronic analysis of the Palaic phonology (e.g., Watkins 1987a and 1987b on lexical aspects, Yakubovich 2005 on a formulaic topos, and Furlan 2007 and Soysal 2016 on the meaning and etymology of specific words). Grammatical overviews have also been published in the last two decades (Melchert 2008b; Kassian and Shatskov 2013), and a comprehensive etymological study is being carried out within the framework of the eDiANA project, which will result in a new digital corpus (https://www.ediana.gwi.uni-muenchen.de).

Most of the studies of Palaic have been very specific and focused on lexical interpretations because of the limited number of documents and lack of bilinguals that could shed light on obscure forms or lexemes. Other issues that have received less attention are no easier to solve. These are essentially sociolinguistic and include a better understanding of the role played by the Palaic culture in the Hittite world; its areal relationship with languages such as Hittite, Hattian, and Luwian; and its status in Hattuša. Was Palaic a spoken language, as suggested by Yakubovich (2010:257, who posits the presence of bilingual Hittite-Palaic scribes in the Hittite capital city) or only a foreign written language? Or was it dead by the time the texts found in the archives of Hatti were written down or copied? While one may argue that the existence of a land of Pala in the earlier manuscripts of the laws (or later copies thereof)⁸ would point to Palaic being spoken there, this kind of circumstantial evidence is only indirect and not particularly trustworthy. Pala might not have been a Palaic-speaking country in the Late Bronze Age, just as Hatti was not, as far as we can tell, a Hattian-speaking kingdom.⁹

The next section provides general information on the classification of Palaic from a linguistic perspective and presents the role and features of the Palaic textual production in Hattuša. This background will be needed to explore the issues raised above.

1.2 The Palaic Language

Genealogically, Palaic is an Indo-European language of the Anatolian branch. Along with (the different varieties of) Luwian and Hittite, it is one of the languages of this branch that are textually attested for the second millennium BCE.¹⁰ Just like Hittite, no traces of Palaic survived the fall of Hatti, but all of the texts that contain Palaic material seem to stem from a relatively early phase.¹¹

The Palaic grammar appears to be reasonably close to that of Hittite and Luwian. Given the limited number of texts, however, it is difficult to establish whether Palaic was closer, morphologically, to the Hittite or Luwic branch. With respect to the leveling of the paradigm of the clitic personal pronouns, Yakubovich (2010:173–178) has shown that Palaic seems to pattern with Hittite and Lydian but not Luwian, with which it does not share the final stages of the innovation. However, the limited number of documents available in Palaic means that negative observations must be made cautiously (e.g., that Palaic seems to be absent in the documents from the kārum in Kaneš, cf. below § 2.1) because Palaic forms may be unrecognizable with our current level of understanding of its lexicon and grammar.

Phonologically, Palaic seems to exhibit the presence of a fricative(?) /f/, which is a very odd feature that does not seem to be matched by any similar phoneme in the other Anatolian languages of the Bronze Age. Although this feature is surprising, the grapheme associated with the alleged phoneme, a sign PI (/wa/, /we/, /wi/, /wu/) with a subscribed vocalic sign, is employed to render Hattian words. Melchert (1994:195) indicates that /f/ is used to write Hattian loanwords, but this characterization is not necessarily entirely correct.¹² All we know is that the Hittite scribes used the same signs to encode what they interpreted as Hattian phonetics in the texts they wrote, regardless of the foreign language of the context. Therefore, writings like la-wa_a-a-at-ta-an-na (KUB 32.18+ iv 2f.), wa_a-a-na (ibid. iv 18), wa_a-a-ar-ra (ibid. iii 12), wa_a-ar-ki-ya (ibid. iv 12; KUB 32.16 iv 6) and wu_ú/pu-la-(a)-ši-na (KUB 35.165 obv. 20 et passim, IBoT 2.36 obv. 5), or again the divine names Katahzifuri—certainly analyzable as Hattian—and the formally elusive Ziparfa/Zaparfa were Hattian forms recognized and rendered as such by the scribes of Hatti or Palaic forms that were mistaken for Hattian ones by the Hittites because of the similar areal origin of the Hattian and Palaic cultures.¹³ Apart from the issue of the aspirate phoneme, which will be discussed again later in this chapter, there exist sound laws that distinguish Palaic from the other Anatolian languages, but, as observed above, we only have a small sample of lexical items to build on.

1.3 The Position of Pala in the Anatolian Historical Geography

The name of the land of Pala is attested in the Hittite Laws, which represent a collection of legal statements¹⁴ from the Old Hittite age (although some copies are composed in a later ductus, and one version of the corpus, the so-called Parallel Text, KBo 6.4, was composed at a later time).¹⁵ The laws do not tell us much about Pala, its location, or its status as a polity:

If anyone kills a merchant, he shall pay 100 minas of silver, and he shall look to his house for it. If it happens in the lands of Luwiya or Pala, he shall pay the 100 minas of silver and also replace his goods. If it happens in the land of Hatti, he shall (also) bring the aforementioned merchant (for burial). (Laws, § 5; Hoffner 1997:19)

While this single passage in the laws has been used to support speculations on the relationship between Hatti and the Palaic-speaking regions in terms of political and administrative subordination and the status of the Palaic peoples during the Old Hittite age, it seems clear to me that all we read is that a different treatment was applied when a felony over which Hatti had jurisdiction was perpetrated outside of the boundaries of the kingdom. As the difference only pertains to the physical possibility or impossibility of retrieving the body of a dead merchant (we do not even know if the reference is to any merchant or a Hittite one!), the passage is not illuminating. It only tells us that Pala was the name of a location in which (Hittite?) traders went to conduct business.

Also, the reference to Pala (like the one to Luwiya, at least in this instance) seems to have been removed from the later parallel text, which characterizes the relevance of Pala for the text as an Old Hittite feature. Contrary to the toponym Luwiya (and Arzawa, which replaces it in one passage of the later copy of the code),¹⁶ Pala does not occur in other passages of the laws. All we learn is that it must have been a location that was geographically or politically distinguished from a Luwian-speaking region and also from Hatti. Whether the mention of Luwiya and Pala in the Hittite Laws indicates that the laws were enforced, as proposed by several scholars (although this has not been supported by external evidence), or whether we are dealing with Hittite cultural propaganda, or again if the reference is to a crime committed in Luwiya/Pala but by Hittites who are then to be punished in Hatti, we do not know, nor will we try to resolve the issue here. That said, the diffusion of Hittite juridical texts outside of Hatti would hint at an intensive circulation of peoples and a Hittite influence outside of Hatti during the Old Hittite era, which would set the scene for early linguistic and cultural phenomena of interference. However, we have already presented evidence for multiculturalism and multilingualism during the Middle Bronze Age, so it will suffice here to state that the hypothesis of intensive interactions between Hatti and an Old Hittite-Palaic polity or geographical area is not inconsistent with the scenario we have defended in the previous chapters. We will not try to use such a complex and elusive source as the laws to try to further support this historical framework.

The earliest documents that contain Palaic linguistic material are roughly contemporary to the age in which we assume that the laws were put into writing: while CTH 751 probably survives in Middle Hittite copies only (as per HPM, contra Carruba 1970:6, 12), CTH 752 survives in an Old Hittite copy or original, as do some unordered fragments with Palaic forms listed under CTH 754.

Neither these texts nor the laws, however, are informative as regards the localization of the land of Pala. The identification of the geographical position of Pala cannot be attempted based on the Old Hittite sources, but only through later documents composed or at least written down during the imperial age.

Diachronically, it makes sense to start with the one that, despite being available only in later versions, was probably copied from an older original. CTH 727, the myth of the Moon fallen from the sky, is preserved in a Hattian and a Hittite version. In the Hattian text, at the very beginning, the spot where the fallen god crashes is the city of Lihzina (Bo 8341:10a, ^[URU]la-ah-za-an zi-ši-im a-ah-ku-un-wa), while, in the Hittite version, the name of the city is omitted and replaced by the common noun KI.LAM (KUB 27.5+ obv. 11, na-aš-kán še-er KI-LAM-ni ma-uš-ta). Lihzina is probably the Anatolian rendering of a Hattian toponym, Lahzan. It could have been a Palaic adaptation as it occurs in the Palaic text KUB 32.18+ (i 14, CTH 752).¹⁷ This dual toponym may point to the existence of a bilingual Hattian/Palaic region or at least of an interface area in which Hattians and Palaeans coexisted. As argued below, this is not inconsistent with the geographical scenario that can be reconstructed.

The other relevant documents were certainly composed and not just copied during the imperial age. These are the Deeds of Šuppiluliuma I and the Annals of Muršili II; both consistently associate the land of Pala with the land of Tummana.¹⁸ Further details may be gained by accepting some identifications of ancient toponyms with modern landscape features. Cammarosano and Marizza (2015) identify the Mount Kaššu mentioned in the Deeds of Šuppiluliuma I (Güterbock 1956a:110; Cammarosano and Marizza 2015:160) with the Ilgaz Dağları, thereby placing Tummana to the north, in the region of classical Blaene (a name assumed to be derived from Pala, just as the name Tummana has been linked to the Hellenistic Domanitis), with Pala being located in the area of modern Ankara. A similar reconstruction using the classical and Hellenistic designations is proposed by Corti 2017a. Corti disagrees with Cammarosano and Marizza mostly on the localization of a third region mentioned in the Late Hittite sources, Durmitta, whose position is, admittedly, problematic, but luckily of little relevance for our present purpose. While Pala can only be roughly located, there are no serious reasons to doubt that it was northwest of the core area of the Hittite kingdom. It is important to notice that toponyms may far outlive political formations and cultural identity groups, so the geographical boundaries of the areas may fluctuate over the centuries. This fact, duly recognized by Cammarosano and Marizza (2015:179), means that it would be unwise to assume that the position of a Pala that was an administrative district of the Hittite Empire from the late 14th century BCE was identical to the location of an elusive polity or geographical region known as Pala, in which the Hittites assumed some crimes could occur that would require rectification and punishment under the alleged jurisdiction of a collection of legal articles conventionally known to us as the Hittite Laws.

1.4 The Areal Context

Uncertainties remain about the exact localization of Pala, and we cannot know for sure for how many centuries into the Late Bronze Age the Palaic language was alive and spoken in the region, but placing Pala roughly to the northwest of Hatti is enough to allow us to attempt some speculations about the areal context of Palaic. The first and most obvious is the contiguity of the Palaic and Hattian speaking regions, which, even though the historical and linguistic geography of Middle and early Late Bronze Age Anatolia is no exact science, seem to have been very close to one another. It would not be surprising if in some regions—for instance, to the south of the Devrez Çayı—both languages coexisted at the end of the third and beginning of the second millennium BCE. This could explain why cities like Lahzan/Lihzina were designated differently in Hattian and Palaic. This observation, while trivial, is particularly important when it comes to discussing the status of the Palaic language in Old Hittite Hattuša. Yakubovich (2010:21, 254–257) finds some similarities in Palaic and Luwian, in that they enjoyed lesser status than the Hattian language and culture, but the cultural compenetration of the Hattian and Palaic elements was probably more relevant to the status of Palaic than its membership in the same language family (Indo-European) and group (Anatolian) as Luwian.

Apart from the obvious areal contiguity with the Hattian world and Hatti, we do not know much about other neighboring cultures and linguistic communities, although some northern regions and populations are mentioned in the later Hittite sources. The two geographical regions that are associated with Pala in the early imperial documentation from the final decades of the 14th century are Tummana and Durmitta.¹⁹ We do not know much about the former, except that it was a buffer kingdom created by the Hittites for military and political reasons. At the time of its creation, it seems to have partly overlapped with the area that we have associated with the southeastern border of Pala. The political and regional makeup of the region must have changed during the previous centuries. Durmitta also appears to have been near what was once Pala, even though much debate exists concerning its location (see also Chapter 4, § 4). If it was, indeed, a northern region, which seems to be the most convincing conclusion,²⁰ it is curious that the ritual of Zuwi, a conjurer who came from this region, contains Luwian sparse Luwian phrases.²¹ The best explanation lies within the relationship that existed between the region of Purušhanda and the northern regions during the Old Assyrian age. Nonetheless, the presence of a large number Luwian speakers in the north in the early years of the Late Bronze Age would be surprising, so that extreme caution is required in evaluating the relevance of this piece of evidence when discussing the areal relationships of the Palaic language (cf. also below § 2.2).

Apart from the political toponyms from the early imperial traditions, the sources also inform us of the presence, in northern Anatolia, of the Kaškeans. This name refers to allegedly nomadic tribes that are assumed to have threatened the stability of the northern regions during the 15th century BCE.²² While speculations have been made about their cultural and linguistic profile,²³ we know nothing of these groups except that they proved hostile to Hatti and resisted entertaining diplomatic relationships with the Hittite kings. Furthermore, the information about them that we can gather from the Hittite sources is relatively late.

2.1 Old Assyrian Age

While Palaic, as previously discussed, textually emerges during the Late Bronze Age in the cuneiform archives of Hattuša, the problem of its areal relationships is rooted in the MBA stages of the Anatolian society during the Old Assyrian phase. The absence of Palaic material from the archives of Kaneš, confirmed for onomastics by the lack of Palaic forms and morphemes in the lexical analysis carried out by Kloekhorst (2019),²⁴ is not surprising, in light of the geographical, historical, and cultural considerations presented in Chapter 4. Given Pala’s presumed localization (see above), it was certainly far too decentered to be involved in the easternmost portions of the Old Assyrian commercial network and, in all likelihood, was not part of it at all. It may have been involved in western interregional networks of which we possess no written records.²⁵ Durhumit, the Assyrian form of the toponym Durmitta, does appear in the Old Assyrian sources but refers to a city relatively far from Kaneš, belonging to a western itinerary. However, since the exact position of Durmitta is far from certain, this does not add much to the reconstruction.²⁶

Linguistically speaking, no recognizable Palaic form has emerged from the documents from Kaneš that have been published, and the typical Palaic diagnostic features are subtle when it comes to identifying a name as Palaic rather than Hittite or Luwian. The task of identification is further complicated by the suboptimal graphemic capability of the Old Assyrian script to render Anatolian forms and clusters. A sound conclusion is that Pala was at the extreme boundary of the area covered by the Old Assyrian trading networks and, if Palaeans were present at all in the kārum society of Kaneš, they probably formed an even smaller minority than the Luwians and left no recognizable trace of their existence in the available documents.

2.2 The Hatti Age

Palaic material is found in the Hittite archives beginning in what is commonly referred to as the Old Hittite phase. This label is nowadays problematic, as the very date of the Old Hittite paleographic ductus has been called into question.²⁷ In this work we maintain with de Martino 2021 that the non-Akkadian texts dating to the reigns of the earlier kings of Hatti probably existed, but it is not possible to establish what the Old Script of the ritual CTH 752 meant in terms of chronology. The text may have been composed at any point between the late Middle Bronze Age and the end of the 16th century BCE.

These chronological uncertainties, however, affect the absolute chronology more than the relative one, so they are not a major issue for our purpose. The data suffice to attribute the production of the earliest documents containing Palaic material to the same phase in which the oldest Hittite documents were composed, such as the first manuscripts of the Laws, the Palace Chronicle, and the Tale of Zalpa.

In this context, which we may label early Late Bronze Age Hatti, Palaic emerges as a foreign language in the Hittite archives. It is associated with Luwian in CTH 752, a text that contains incantations in both languages, which means that the two idioms shared the status of cultic vernaculars, although it does not prove that they had the same role and prevalence. The fact that Pala and the Luwian land are both mentioned in the passage of the Laws discussed above indicates that the two areas (and populations) had a similar political status from the perspective of the Hittites. However, the Luwian region is mentioned in other loci of the code, whereas Pala occurs only once and disappears in the later version of the text.²⁸

From a linguistic perspective, there are no obvious traces of significant linguistic interference or interaction between Luwian and Palaic. The shared lexical elements, such as the title tabarna/labarna, that also occur in Hattian and Hittite depend on a mixed cultural and linguistic environment that does not prove direct interference between Palaic and Luwian at any known stage of the history of Anatolia.

Beyond the lexical level, Palaic does not share the same sociolinguistic leveling of the clitic pronominal system that was identified for Luwian by Yakubovich (2010:161–195), nor did it develop the i-mutation, which, while conditioned by prosodic factors at least in Lydian (cf. Sasseville 2017), was spread areally through the influence of Luwian (or Luwic), at least in the case of Hittite.²⁹

Despite being attested only in the Hattuša archives, Palaic does not show any trace of active linguistic interference with Hittite either. Unlike Luwian, that will become one of the main vernaculars in the Hittite regions of Anatolia and in the capital city by the pre-imperial age, the language of Pala seems to have existed only as a cultic language and to have played no recognizable role in the development of the Late Hittite language. Whereas Luwian had lexical, morphological, and even morphosyntactic influences on Late Hittite, no changes in Hittite were induced by contacts with Palaic. And, as just observed regarding Luwian-Palaic shared lexical items, the only words that occur in both Palaic and Hittite contexts (tabarna/labarna being the obvious example), are common areal words.³⁰

Therefore, Goedegebuure’s observation (2008, 170) that the only attested interference is between Palaic and Hattian seems trustworthy. The evidence for interference can, as usual, be divided into two main types: lexical phenomena and grammatical phenomena. In the next section, the available evidence will be critically discussed.

2.3 Alleged Phenomena of Linguistic Interference between Hattian and Palaic

Hattian-Palaic lexical interference seems proven by the existence of a limited group of Hattian words in the Palaic incantations, ritual passages, and texts from Hattuša. The candidate forms, which are based on the Carruba’s glossary (1970), are mostly those that were already listed at § 1.2: lafattanna (la-wa_a-a-at-ta-an-na),³¹ the mysterious compound(?) manzakilba- (ma-an-za-ki-⸢il⸣-ba-),³² tuwafanteli (tu-wa-wa_a-an-te-li),³³ fana (wa_a-a-na),³⁴ farra (wa_a-a-ar-ra),³⁵ farkiya (wa_a-ar-ki-ya),³⁶ fašhullatia (wa_a-(a)-aš-hu-(ul)-la-ti-ya-aš), fatia (wa_a-ti-ya-),³⁷ fatila (wa_a-ti-la-),³⁸ and fuzzanni (wu_ú-(uz)-za-(an)-ni-),³⁹ all of which are semantically unclear but occur in the usual cultic and ritual contexts; possibly the title taberna/labarna (if originally Hattian);⁴⁰ and the bread fulašina (wu_ú/pu-la-(a)-ši-na,⁴¹ also attested in the derivative wu_ú-la-ši-ni-ki-eš). Two observations are in order. First of all, recent advancements in our understanding of Palaic suggest that some of the words containing the putative fricative sound may have been inherited rather than borrowed (cf. Sasseville 2020:368–369), a circumstance for which one mayaccount either by reconstructing a dedicated Palaic phoneme or, perhaps more reasonably, by assuming an interpretation of the words as belonging to a common Palaic-Hattian lexicon familiar to the Hittite scribes, who rendered them using the same graphemic device. Second, regardless of the true origin of each word, based on the contexts (and on the meaning in the case of the fulašina bread), the lexical items belonged to the technical lexicon of the ritual sphere, which may indicate that they were composed by scribes who specialized in writing ritual texts and possibly employed a standard graphic inventory. This observation is conducive to a further methodological remark: since we know only the written Palaic (and Hattian) grapholect of Hattuša, it is impossible to detail the exact nature and duration of Hattian-Palaic interference. Even the idea that Palaic had a fricative phoneme (either inherited or used to adapt borrowings) is, despite the apparent evidence, mere speculation about a language of which we have no direct examples. The texts that we possess were filtered by at least one and possibly two intermediate cultures.

Possible traces of structural interference, however, have been proposed to exist. The contrastive particle -pi of Palaic seems to functionally resemble the -pa of Luwian and the -(m)a of Hittite, and is formally remindful of the contrastive/adversative particle -pi of Hattian.⁴² The hypothesis that the Palaic particle was borrowed by Hattian was cursorily advanced by Goedegebuure (2008:170–171). While this is theoretically possible, the borrowing of a clitic grammatical morpheme and its integration in the morphosyntax of the target language (in this case, the Hattian pi would be integrated as a Wackernagel particle in the syntagmatic structure of Palaic) is a phenomenon that occurs in contexts of very intensive contact. Assuming it before excluding a simple Indo-European inheritance is acceptable only if a very convincing match exists not only formally but also semantically. Given our limited lexical understanding of the Palaic and Hattian languages, claiming that the function of either of the particles was certain would be unfounded. Therefore, we are once again faced with circumstantial and non-conclusive evidence.

Another possible example of interference would be the morphophonemic rule of assimilation of the alveolar nasal /n/ to /m/ before a bilabial stop.⁴³ This seems to regularly occur in Hattian, judging from the texts we possess, as well as in Palaic, but far more rarely in the Hittite or Luwian texts produced in Hattuša. It is very difficult to evaluate this possible case of morphophonemic interference because the texts that we can read in both Hattian and Palaic were composed in Hattuša and reflect the way that these languages were perceived and rendered (possibly through dictation) by the Hittite scribes. However, given that this common pattern does contrast Hattian and Palaic with Luwian and Hittite, the hypothesis that the feature was a northern areal one cannot be dismissed entirely.

Summing up, one can only admit that the evidence for areal contacts involving Palaic is meager in the limited number of texts available to us. Nevertheless, even if none of the features that Hattian and Palaic share can be called conclusive, each could point to language superposition or attrition due to the cohabitation of Hattian and Palaic speakers in a geographically contiguous and partially overlapping territory. This scenario appears to be consistent with the strongest, although not properly linguistic, piece of evidence supporting a strong connection between the two cultures, which is the composition of a putative ‘Palaic’ pantheon as we can reconstruct it based on the sources that carry Palaic texts. There are two types of divine names: Hattian names (e.g., ^dHašamili and possibly also ^dZiparfa/Zaparfa), and Anatolian names, occasionally with a Palaic suffix (e.g., ^dIlaliyantikeš and ^dUliliyantikeš). Even though proper names, including names of divinities, are linguistically very conservative and therefore may return a palimpsest of which the strata are impenetrable, the evidence is certainly compatible with a scenario of strong cultural and linguistic contacts involving the Hattian, Hittite, and Palaic worlds.