Ancient Macedonian Language - a distinct Indo-European language

Ancient Macedonian Language - a distinct Indo-European language

Macedonia and Greece: Two Ancient and Separate Nations - John Shea 1997 pp.23-35

To get to the real Macedonians we need to start a little before the time of Alexander the Great. If we go too far back, say to the seventh century B.C., we find that Macedon was a tiny little piece of land that no one today would really be interested in. It was an area that could be covered on horseback in a day's ride. Macedon at first included the area immediately east of Lake Kastoria and east and north of the Haliakmon River. Certainly there is little glory to claim from this period of Macedonian history. By the fifth century B.C. the kingdom had been extended eastward to what is now the Struma River, and a century later the Macedonian homeland was extended to include all of the territory West of the Nestos River.' In the time of Philip II and his son, Alexander the Great, the Macedonian homeland was at its largest, and Macedonian power was at its peak. This seems the obvious era in which to begin our enquiry.

Modern Greeks prefer to think of the ancient Macedonians as Greeks. This was part of their justification for taking a part of Macedonia by conquest earlier in this century, and is still used to justify their present international position. Greek arguments frequently focus on the time of Alexander because of his undoubted influence in spreading Hellenic culture to distant parts of the known world. It is clear, too, that they gain some satisfaction from imagining some family connection with that extraordinary figure. However, the modern Greek ideas would have been rejected by both the ancient Macedonians and the ancient Greeks.

If we start by looking at modern Greek discussions of these ideas we can then consider what historians have to say about their arguments, point by point. We get some of the flavor of Greek attitudes in the Greek publication Macedonia, History and Politics (published by George Christopoulos, John Bastias, printed by Ekdotike Athenon S.A. for the Center for Macedonians Abroad, and the Society for Macedonian Studies, 1991). This is a publication available in Greek embassies and distributed to Greek communities and multi-cultural organizations throughout the English-speaking world. The author of this book considers that the use of the Greek language by Macedonians is proof of their Greekness. In passing we might reflect on the modern use of English by many countries as a convenience for trade or war, and note that this usage proves nothing at all about the ethnicity or culture of the users. However, the author of Macedonia, History and Politics claims that the dissemination of the "Greek language and Greek culture throughout the known world by Alexander the Great and his Macedonians provides the most irrefutable confirmation'' of the unity of the Macedonians with the other Greeks.

To explore thoroughly this issue of the proposed Greekness of the Macedonians, we need to consider evidence from a number of quarters. If the early Macedonians were Greek you would expect that (a) there might be clear evidence that the language of the Macedonians was a dialect of Greek, rather than a separate branch of the Indo-European language group; (b) writers of the time would have recognized Macedonians as Greek rather than as foreigners and would have spoken about Macedonia as though it was a part of Hellenism; and (c) historians today would speak of the ancient Macedonians as though they were Greek in ancient times. As we will see, none of these ideas is unequivocally supported.

Linguistic Evidence

In questioning the significance of the use of Greek by the ancient Macedonians we need to sort out some of the linguistic history of the Macedonians. Firstly, the language of the original Macedonians, whatever it was, existed long before Macedonia became a powerful state. This is before the time of the great kings Philip II and Alexander the Great. The name "Macedones" originated many centuries earlier, and probably came from the "real" Macedonian language. If the Macedonian language was recognized as Greek, and understood by Greeks, you would expect that this was the language used by the great Macedonian kings in a formal or legal context. But it was not.

We know with some certainty that Attic Greek, which came from much farther south (around the Athens area) and was being used in other parts of the world as a trade language, was used more and more as the language of state and used also in Alexander's multi-cultural army. No linguist accepts that this language was the original Macedonian. So we have clear evidence that the Greek used by the Macedonians was a new language. Therefore one cannot argue that the use of this language proves any linguistic associations between the original Macedonians and Greeks.

Many scholars have concluded that the ancient Macedonian language was not a Greek dialect and that it was more or less related to the languages of Macedonia's northern neighbors, the Illyrians and the Thracians. These scholars include Muller and Mayer, writing in the nineteenth century, and Thumb, Thumb-Kieckers, Vasmer, Kacarov, Beshevljev, Budimir, Pisani, Russu, Baric, Poghirc, Chantraine, Katicic, and Nerosnak, writing in the twentieth. Here attention will be given to sources more readily accessible to those who want to inquire further.

The problem for modern-day linguists is that not a single sentence of the original Macedonian language has been retained. All that is left are records of proper names and isolated words -which, as historian E. Badian of Harvard University points out, is hardly sufficient basis for judgments about linguistic affinities.' We do know that the Macedonians increasingly came to use a southern form of Greek in their formal dealings. Traian Stoijanovich tells US3 that in the fifth century B.C., the Macedonian rulers abandoned Macedonian and began using Attic Greek for public administration. This did not change the attitudes of the Greeks, who still regarded the Macedonians as barbarians.

However, Stoijanovich says it is not known whether the ancient Macedonian language was an independent language or a Greek dialect into which a non-Hellenic vocabulary and certain other non-Hellenic traits were introduced. Like other historians, he considers it quite possible that Macedonian was the language of the ruling class and that a considerable proportion of the subjects of the Macedonian chiefs spoke other languages.

Peter Hill, author of the section "Macedonians" in the official Australian bicentennial encyclopedia, The Australian People (perhaps 200,000 Macedonians live in Australia), writes:

What is certain is that Alexander's mother tongue was not Greek. Alexander enjoyed a Greek education and adopted Greek as the language of his empire but to claim that that made him Greek is to suggest that the Irish and the Indians are really British because they have adopted English for administrative purposes.

Like Hill, E. Badian refutes the assumptions that a nation is essentially defined by a language and that a common language implies a common nationhood. He argues that this latter idea is patently untrue for the greater part of human history and to a large extent even today. The formal written language of ancient Macedonians was inevitably Greek, as was the case for various other ancient peoples. There was really no alternative. However, this in no way assures good relations between peoples, nor does it necessarily show any consciousness of a common interest. What is of greater historical interest, Badian says, is the documented evidence that Greeks and Macedonians regarded each other as foreign.

The use of the Macedonian language by Alexander's infantry. The Macedonian kings, Philip and Alexander, favored Hellenization and encouraged the use of Attic Greek in their administrations, but the use of this foreign tongue was not foisted upon ordinary Macedonians. Although at least some of Alexander's Greek companions knew the Macedonian language, having come to Macedonia at an early age, Alexander never tried to impose Greek on his Macedonian infantry or to integrate this infantry with Greek units or Greek "foreign" individuals. Alexander's infantry continued to use the Macedonian tongue even late into his Asian expeditions. Badian describes some convincing cases in which Macedonian troops could not follow commands in Greek. For instance, during his argument with Clitus, which led to his good friend's death, at the end Alexander is said to have called for his guards in Macedonian when he felt his life threatened. Badian rejects the idea that this was a reversion to a more primitive part of his psyche, under stress. He prefers the simpler explanation that Alexander used the only language in which his guards could be addressed.

To establish his case, Badian quotes a surviving papyrus fragment that seems to be the only good source to reveal the facts of the infantry use of Macedonian. This fragment tells of a battle, early in 321 B.C., in which the Greek commander Ambiance faced the Macedonian Neoptolemus with his Macedonian phalanx. Wanting to have the Macedonians join him rather than fight him, Ambiance needed to convince them of his superior position. The story continues:

When Eumenues saw the close-locked formation of the Macedonian phalanx ... he sent Xennias once more, a man whose speech was Macedonian, bidding him declare that he would not fight them frontally but would follow them with his cavalry and units of light troops and bar them from provisions.

Badian tells us that Xennias' name reveals him to be a Macedonian. Since he was with Ambiance he was probably a Macedonian of superior status who spoke both standard Greek and his native language. Ambiance needed this interpreter to transmit his message. This means that the phalanx had to be addressed in Macedonian if they were going to understand. Ambiance did not address them himself, although this was the common way for leaders of the time, nor did he send a Greek. Badian concludes that Greek was a foreign tongue to the Macedonians. Similarly, Alexander used Macedonian to address his guards because it was their normal language, and he had to be sure he would be understood. It also seems clear that educated Greeks did not speak the Macedonian language unless (presumably) they had grown up with Macedonians and learned it from their childhood friends, as some of Alexander's Greek companions must have.

Other facts are consistent with this argument. Philip II seems not to have used any Greek commanders for his Macedonian troops. Presumably, the first generation Greek immigrants into his cities had not learned the language. It is also a fact that Ambiance, the commander in the story above, was notorious for the trouble he repeatedly had in getting Macedonian infantry to fight for him, even though he was an able leader. His problem was probably not simply his troops' antagonism to the fact that he was Greek. His problem was that he could not directly communicate with Macedonian soldiers. In the end this defect cost him his life.

Political reasons for the use of the Greek language. Considering the use of Greek as the language of command in Alexander's armies, R.A. Crossland concludes that this development was a matter of administrative efficiency. Although it was the Macedonians who had to learn Greek at first, the same requirement was made of at least some of his Persian troops after many conquests. For a long while Alexander thought that Greek was the best language to use as the common medium of communication among the peoples of his empire, "and not because Macedonian was similar to it." Nevertheless, as we have already noted, even by the latter part of his Asian campaigns, Alexander's infantry still did not speak a Greek language.

In other words, a very important reason for Hellenization of the Macedonians was their new role of political power-broker. The Greek language was available in written form and was widely used throughout the Macedonian sphere of influence. It was a very convenient vehicle for use in creating an international empire, something that both Philip and Alexander hoped to do. Its use may have also have led to some appeasement of Greek hostility towards the dominating Macedonians. All of these are sound reasons for choosing to use the Greek language as the tongue of administration throughout the expanding empire. However, after a time the value of Greek culture to the Macedonians' cause began to fade. Eventually Alexander began to think in terms of a blending of the diverse cultures of his great empire. Perhaps in order to appease his new Persian subjects, it was now the blending of Macedonian and Persian that mattered, rather than the blending of Macedonian and Greek.

Macedonian attitudes to the Greek language. For the most part we have little information on Macedonian attitudes to the Greeks or their language. Badian reminds us that no Macedonian oratory survives, since the language was never a literary one. However, he concludes that the existence on both sides of a feeling that they were "peoples of non-kindred race" is very probable. The language barrier would keep this awareness alive, even though the literary language of educated Macedonians could only be Greek. That fact was as irrelevant to ordinary people, and perhaps even to those of higher status, as was the Hellenization of the Macedonian upper class. Badian gives a more recent example of a similar phenomenon. In eighteenth-century Europe, French language and culture prevailed amongst people of education. In fact, during the early part of the eighteenth century the language and culture of the German royal courts, including that of Frederick the Great in Prussia, were French. Most of the books published in Germany in the first half of the century were in Latin and French! Thus upper-class German ladies might write only in French, yet this did not mean that they were French or even Francophile. Badian suggests that Clitus' anger toward Alexander was representative of a persisting antagonism to Greeks and their ways seen among all classes of Macedonians. He says that these feelings are most clearly evident where the historical record deals with ordinary people, like the Macedonian infantrymen referred to above.

The linguistic character of ancient Macedonia. Arnold Toynbee asserts that the Macedonians of all ancient historical periods spoke Greek. He argues firstly that "they (the Makedones) were already Greek speaking 150 years to 200 years earlier than Augustus' time." This observation would seem to be of little weight in the present discussion since we have already noted the increasing, and deliberately chosen, use of Attic Greek by the Macedonian nobility. The use of a language from a distant location by a limited number of noble families tells us nothing about the native tongue of the Macedonians of the fourth century B.C., the Anglo-Saxons of thirteenth century England, or the Prussians of early eighteenth century Germany.

Nevertheless it is worth looking at Toynbee's point a little further to uncover its internal inconsistencies. Toynbee describes an occasion in 167 B.C. when L. Aemilius Paulus announced in a public speech at Amphipolis the Roman government's decisions for the settlement of continental European Greece. This speech was delivered in Latin, but there was a Greek translation of the speech "for the benefit of Paulus' audience which was drawn from all parts of Greece." From this Toynbee concludes that at this stage the Macedonians were Greek-speaking, since in the public meeting place at Amphipolis, the majority of the listeners must have been Macedonians. Yet Toynbee himself states that the Greek translation was provided because the audience "was drawn from all parts of Greece." However, if we follow Toynbee's line that we are dealing with a diverse group of native Greek speakers, many of whom were Macedonian and who, according to Toynbee, spoke a dialect of Greek that no other Greeks could understand, it is asking a bit much to expect us to believe that these representatives suddenly all understood the same "Greek"- that is, unless the "Greele' that was used was the koine, the international version of Greek developed from Attic, that was widely spoken in this area of the empire at the time. The audience was made up largely of leaders of one kind or another, people who were most likely to speak such a language. It is likely that virtually any trader, businessman, administrator, or political leader of the time would have spoken this language (or would have been in the company of an interpreter who could), as well as his own vernacular and perhaps other trade or administrative languages as well. Thus the translation of Paulus' speech into Greek tells us absolutely nothing about the native language of the Macedonians or of anyone else.

Toynbee presents other arguments based on linguistic analysis to support his contention that the Macedonians were native Greek speakers. He asserts that Macedonian is Greek based on the "Greekness" of the word "Makedones" and its variant "Makednoi," Macedonian place names, the names of the members of the Argead house, all recorded Macedonian personal names, the names of Macedonian from Upper Macedonia, the names of the Upper Macedonian cantons, the names of the Macedonian months, the majority of which he claims as Greek. Though at first glance this kind of analysis seems weighty, the counter-arguments are at least as powerful.

An issue that we have to deal with here is what constitutes a "Greek name." It is generally accepted that Indo-European Greeks, Illyrians, Thracians and others settled in the Balkan Peninsula in the fourth, third, and second millennium B.C. As we will see later in more detail, it has been argued that only 40 to 50 percent of the vocabulary of Greek is Indo-European in origin and that 80 percent of its proper names cannot be explained as Indo-European.9 At least two possibilities might explain the presence of such linguistic forms in ancient Greek. One is that pre-Hellenic cultures were non-Indo-European and that the Greek newcomers adopted many proper names and other words from those peoples. Alternatively, the words might have been introduced by conquerors and settlers from the Levant and from Egypt in the second millennium B.C. In either case it is quite possible that such words came into Macedonian and other Balkan languages in the very same way. Thus both languages might have borrowed from others. If we favor the modern view that the pre-Hellenic influences in Greek are non-Indo-European, and we take into account the observed fact that place names often tend to last through conquest and assimilation, its would be reasonable to assume that some of the supposed "Greek" place names found in the "Macedonian" language are in fact pre-Hellenic names.

It is easy to find modern examples of the same phenomenon. Both France and Germany have many Celtic place names yet do not speak a Celtic language, or even the same language. The people of England are "British," a name based on a Latin word formerly applied to a Celtic-speaking people and now referring to an Anglo-Saxon people. A study of the word "British" does not help us to determine what language the British speak. It is certainly not Latin, yet there is historical evidence about the use of Latin in Britain, the same kind of evidence that is trotted out to prove that the Macedonians were Greek. For instance, since English coins have Latin on them, we might conclude that the British speak Latin, following the argument that it would not make sense to use a language no one could read on such common items. Similarly, many English parish churches have collections of epitaphs in Latin, dating from the Middle Ages. Classicist Andy Fear points out that most of the population of medieval England could not even read English, let alone Latin. Obviously, the significance of surviving Greek texts from Macedonia must be treated with caution. Fear notes, too, that Greek inscriptions from ancient Macedonia are in a mixture of Greek dialects. It is much easier to believe that this could occur if Greek was alien to Macedonia, instead of the common language. If the latter were the case, we might expect to see a consistent form employed.

If we study the month names used in England and France, we can see that they resemble each other. This is not a basis for concluding that French and English are the same language. All one can reasonably conclude is that there has been similar heavy influence across these two languages. To say, for such superficial reasons, that Greek and Macedonian are the same language is to make far too much of a little thing. We must remember also that much of the history about ancient Macedonians that is passed on to us comes through Greek sources, and names are likely to have been shaped into Greek forms for a myriad of reasons, including the likelihood that Greek writers may not have been able to pronounce other tongues. A modern analogy would be to think that France is a German-speaking country because when reading a German textbook one comes across the name "Frankreich" ruled by, say, Karl rather than Charles. It is easy enough to find English forms of foreign place names that look far removed from their native form; Florence for Firenze, and so on.

In his essay "Linguistic Problems of the Balkan Area in Late Prehistoric and Early Classical Periods,"o R.A. Crossland directly addresses the issue of the linguistic character of ancient Macedonian. Crossland points out that the principal languages of the Balkan region in question* appear to have been Illyrian or an Illyrian language group; Thracian or Thraco-Dacian; and Macedonian. When it comes to the language of the Macedonians, Crossland takes a position very different from modern Greek writers. He rejects the idea that the Macedonians and their language were of Mycenaean origin. Then he goes on to consider linguistic and archeological evidence about the possible origins of Macedonian and in so doing directly contradicts Toynbee.

Crossland points out that the territory of the Macedones at the beginning of the fifth century B.C. seems to have lain between Tymphaea in the west, Pelagonia in the north and the river Axius in the east, but so far no category of place-names that we can identify as "Macedonian" has been identified in this area, and no inscription in Greek earlier than the late fourth century B.C. has been found in any part of Macedonia. Thus we have no substantial evidence about the nature of the Macedonian language in the time that it was most exclusively used (before the fifth century B.C.), but neither do we have evidence of any Greek language being in use at that point in history. The use of Greek came later.

Crossland says that the names of Macedonians mentioned in fifth- and fourth-century sources are almost all either certainly or possibly Greek, but he argues that this is not significant, since members of one people often borrow names from another whom they regard as culturally superior. Certainly the Macedonian craze for things Greek, including Greek education for the children of the upper classes, suggests such an attitude.

Next, Crossland points out that the ancient writers of the time gave imprecise information about the language of the Macedonians. None of the ancient Greek writers gives a detailed statement about the language that the Macedones spoke. The limited evidence that remains consists of words preserved by Greek lexicographers, especially Hesychius, from about the fifth century A.D. According to Crossland, these words were listed as "used by the Macedonians" or "used in Macedonia" without any indication of the origins of the words. Crossland also cites several other authorities who confirm his conclusions.

Regarding the ancient writers' capacity to recognize significant linguistic features, Crossland agrees with Toynbee in pointing out that when language and speech seemed very different the ancient writers might have had difficulty in making correct classifications. We do not have an understanding of the details of their systems for classifying language. However, we need to remember that only in very recent times have linguists recognized the many languages that make up the Indo-European group. Crossland says that it is difficult to know whether one group of Greek speakers, say the Athenians, would have been able to recognize really different dialects of Greek, or whether they would have been influenced by differences of culture to classify such dialects as barbarian.

Crossland says that the evidence available is too sparse and unsatisfactory to tell us conclusively whether Macedonian was a dialect of Greek or a distinct language. He notes that another authority, N. Hammond, has actually concluded that Macedonian was a dialect of Greek, based on interpretations of information in ancient sources about the status and use of Macedonian under Alexander the Great and his successors. However, Crossland is skeptical of Hammond's reasoning and says that better evidence would come from comparative linguistic study.

Crossland says that two kinds of evidence would help us to conclude that Macedonian was a dialect of Greek. Firstly, we would have to be able to observe or reconstruct its sound system and morphology in a way that would reveal any similarities to recognized ancient Greek dialects, and any contrasts to other Indo-European languages. Secondly, we would have to know whether speakers of most of those Greek dialects could understand and be understood by Macedonians. But none of the necessary evidence is available. The lexical items thought to be Macedonian are too few and uncertain for any useful reconstructions of the language's sound system or morphology, and no Greek writer of the fifth or fourth century B.C. states explicitly whether Greek speakers such as the Athenians could understand the native speech of the Macedonians. Crossland says that these Greeks seemed to have had no difficulty in communicating with the Macedonian court, but this is probably because the royal family of Macedonia, and perhaps most of the nobility, spoke Attic Greek fluently. At home with their families or with their own clansmen they probably used their native tongue, Crossland believes.

We do not know either what form of "international" Greek speech might have been used in Macedonia since there are no substantial inscriptions in Greek from Macedonia earlier than the third century. The Greek speech used might have been Attic or an early form of the koine deriving from it that was already spoken even more widely in the Balkans before Alexander's conquest of the Persian Empire.

The information about supposedly Macedonian words given by ancient lexicographers may not be very reliable. Along with words that were a part of the real Macedonian tongue in the fourth century B.C., they might have listed words and usages typical of the variety of Greek that was used in Macedonia from the third century onwards. They may also have included words that were special to the Macedonian armies. Some Greeks in the early Hellenistic period may even have regarded as Macedonian words that belonged to the koine as a whole, but not to Attic. We have no way of knowing the underlying basis for classifying words as belonging to one language or another.

Crossland is very critical of Kalleris, a Greek writer who tries to make a case from a linguistics standpoint for Macedonian being a Greek dialect. It is worth looking at this material in detail because of its apparent thoroughness, and because of its relevance to Toynbee's arguments.

In an examination of the 153 words that are described as Macedonian in ancient sources, Kalleris considers that well over three-quarters of these words are Greek. Crossland finds this quite unconvincing. First, he says, a third of these words have no satisfactory etymology. Second, he says that a further 44 items should be disregarded as being false forms in the sources from which they came. They are simply adjectives of Greek formation based on place-names. Although these words seem to be Indo-European, they could belong to an Indo-European language other than Greek. Some of them might be military or technical terms which are Attic in form and were borrowed from Attic Greek in the fifth or fourth century.

Third, Crossland argues, if Macedonian was a dialect of Greek it is extremely unlikely that it would have been similar to Attic Greek. The original Macedonians did not come from the area of Athens and share no history with the Athenians. This means that the Attic words are a false lead, just late borrowings from Greek. It would be much more convincing, perhaps crucial, to find Macedonian words that were not specifically Attic but which occurred either in a considerable number of Greek dialects or in some of the dialects that were spoken in areas adjacent to Macedonia. Kalleris gives fifty-one words of this kind. Many of these words occur in Doric or other West Greek dialects or resemble words in these dialects. However, it is quite possible that these words were borrowed from West Greek dialects or from Thessalian, particularly since all except eighteen of them are the sort of words which the Macedonians might well have borrowed from their neighbors. They include titles of gods, names of festivals and months of the year, military terms, and names of objects that they might have learnt from neighbors to make and use. Such words are often borrowed from neighboring groups, so their existence in Macedonia is not convincing evidence that they were originally Macedonian.

Fourth, the remaining eighteen words, none of which corresponds exactly in meaning or form with Greek words, seem insufficient to make a case for classifying Macedonian as Greek. Once again there is the possibility that the words were borrowed from neighbors. At the western and southern borders of Macedonia were tribes speaking different Greek dialects, and we know that the Macedonians were in contact with these peoples. The Thessalians to the south are particularly likely to have been influential since they were culturally and politically more advanced than the Macedonians before the fifth century. They are likely to have influenced the Macedonians particularly strongly until the growth of Athenian influence. Herodotus reports on traditions in the same period of close contact between the Macedonians and the Dorians before the latter were supposed to have migrated southward.

Finally, though again it is hardly sufficient basis for any conclusion, there is one language feature evident in the surviving "Macedonian" words that points to the idea of a separate language. Macedonian seems to have had a phonological feature that marks it as different from Greek dialects. This is the correspondence of a sound written with B, to Ph in Greek. For instance, this would appear as something like Bilippos in Macedonian, and Philippos in Greek. Crossland says that this change puts Macedonian closer in phonology to Illyrian and Thracian than to Greek, but it does not mean that Macedonian was a dialect of either language.

Crossland is not convinced by claims that comments from writers such as Arrian and Plutarch in the first to second centuries A.D. (e.g. Plutarch, Ant. 27) show that Macedonians spoke a dialect of Greek as their native tongue. He says they are inconclusive since the expressions used are vague and might be referring to a "Macedonian style" rather than a "Macedonian language" or "dialect." These descriptions would be just as likely if Macedonian was a distinct language as they would be if it was a dialect of Greek. Crossland points out that it is possible that Macedonian kings and their courts, soldiers and colonists might have continued to speak a second language in their homes and among themselves for some generations even though they spoke Greek for most practical purposes. After all, it is easy to think of examples of this kind of thing in more modern times. Crossland notes that Gaelic was used alongside English for generations by Scots who emigrated to America. It is still used in this way in some small communities in North America. Similarly, although English was used as the language of command and administration in British army regiments recruited predominantly in Wales, the Welsh language was still used privately.

Like historians who have examined this question, Crossland suggests that Alexander may have required Macedonians in his armies to use Greek as the language of command, just as he required many Persians to learn it (Plut. Alex. 43.7), because it was efficient, and because he thought it the language best suited to serve as the common medium of communication among the peoples of his empire. This kind of strategic decision does not require that Macedonian should have been similar to the new "international" language.

In summing up, Crossland says again that the evidence does not indicate convincingly that Macedonian was a dialect of Greek rather than a separate Indo-European language. Even Toynbee, who is persuaded in the opposite direction by the very flimsy evidence we have considered above emphasizes that the evidence is "fragmentary, ... confused and self-contradictory." In practical terms this suggests that modern Greeks may have to look elsewhere for convincing evidence that ancient Macedonians were Greek.