For administrative and political purposes, Attic Greek seems to have operated as a lingua franca amongst the ethno-linguistically diverse communities of Macedonia and the north Aegean region, creating a diglossic linguistic area. Attic Greek was standardized as the language of the court, formal discourse and diplomacy from at least the time of Archelaus at the end of the 5th century BC. Attic was further spread by Macedonia's conquests. Although Macedonian continued to be spoken well into Antigonid times, Attic became the prevalent oral dialect not only in Macedonia, but throughout the Macedonian-ruled Hellenistic world
Attempts to classify Ancient Macedonian are made difficult by the paucity of surviving Ancient Macedonian texts, as it was a primarily oral language and most archeological inscriptions indicate that there was no dominant written language in Macedonia other than Attic, and later Koine Greek. All surviving epigraphical evidence from grave markers to public inscriptions is in Greek.
Classification attempts are based on a vocabulary of 150-200 words and 200 personal names assembled mainly from the 5th century lexicon of Hesychius of Alexandria, as well as a few fragmentary surviving inscriptions, coins, and the occasional passage in ancient sources. Most of the vocabulary is regular Greek, with tendencies toward Doric Greek and Aeolic Greek
; on the other hand, there can be found some Illyrian and Thracian elements
. The Pella curse tablet
, which was found in 1986 at Pella and dates to the mid-4th century BC or slightly earlier, is believed to be the only substantial attested text in Macedonian
. The language of the tablet is a harsh but a distinctly recognizable form of Northwest Greek: the tablet, therefore, has been used to support the argument that ancient Macedonian was a Northwest Greek dialect and mainly a Doric dialect
, whilst Hatzopoulos's analysis revealed some tendencies toward the Aeolic Greek dialect
. Macedonian onomastics paint a similar picture, most personal names being recognizably Greek
(e.g. Alexandros, Philippos, Dionysios, Apollonios, Demetrios), with some dating back to Homeric (e.g. Ptolemaeos) or even Mycenean times, though here too there can be found the occasional non-Greek name
(e.g. "Bithys"). Macedonian toponyms and hydronyms are similarly overwhelmingly Greek in origin
(e.g. Aegae, Dion, Pieria, Haliacmon), as are the names of the months of the Macedonian calendar and the names of most of the deities the Macedonians worshiped
, and according to Hammond, these are not late borrowings
. Nevertheless, a definitive conclusion eludes the linguistic community. On the one hand, Macedonian shares close structural and lexical affinity with the "proper" Greek dialects (especially Northwest Greek and Thessalian)
. The majority of the words are Greek, although some of these could represent loans or cognate forms. On the other hand, a number of phonological, lexical and onomastic features also set Macedonian apart
. These latter features, possibly representing traces of a substratal language
, occur in what are considered to be particularly conservative systems of the language.
Several hypotheses have consequently been proposed as to the overall "position" of Macedonian, but all broadly see it either as a peripheral Greek dialect, a separate yet related language (see Hellenic languages), or even a hybridized idiom. Drawing on the similarities between Macedonian, Greek and Brygian, several scholars suggest that they formed an Indo-European macro-dialectical group which split before circa 14th-13th century BCE (i.e. prior to the appearance of the main Greek dialects). The same data has been analyzed in alternative manner, which sees the formation of the main Greek dialects as a later convergence of related but distinct groups. Macedonian did not fully participate in this process, making its ultimate position difficult to define other than being a contiguous, related 'minor' language.
Another stream of evidence is metalinguistics and the question of mutual intelligibility. The available literary evidence cannot provide detail as to the exact nature of Macedonian, however, it does suggest that Macedonian and Greek were sufficiently different to pose communication difficulties between Greek and Macedonian contingents, even necessitating the use of interpreters as late as the time of Alexander the Great
. Based on this evidence, Papazoglou has argued that, by definition, Macedonian could not have been a Greek dialect. However, similar evidence for non-intelligibility exists for other ancient Greek dialects, e.g. Aetolian