Herodotus and language.
At first glance, Herodotus does not seem to be particularly interested in different languages. He is often criticised for neglecting to mention the presence of interpreters and the problems associated with translation (Harrison 1998:3; Waters 1985:79; Robinson 2002:1). The general Greek attitude towards language is clear from the fact that language was the feature they used to distinguish themselves from non-Greeks (Harrison 1998:1). They called them [TEXT NOT REPRODUCIBLE IN ASCII.] (Robins 1967:11), "the stammering ones", but the term came to mean "uncivilised", and has retained this meaning in the English word barbarian,, while the linguistic connotation has been lost. One might think, at first, that this exemplifies Herodotus' view of language; that he sees Greek as the only "real" language and, therefore, that discussions about the details of language and the intricacies of translation are not worth mentioning. But as one looks deeper into what Herodotus has to say, it seems that he did in fact show a certain degree of interest in different languages and multilingualism. He has provided a sophisticated description of certain sociolinguistic phenomena, of dialects and of language families. He also describes a number of examples of second language learning by both children and adults and hints at a few of his personal beliefs about the nature of language. He also, despite the criticism levelled against him, gives us some insight into and recognition of the issues of translation and interpretation.
Herodotus refers explicitly to different languages and language-related issues a number of times in his work. It seems that he could not speak any language other than Greek. While Harrison (1998:1-2) entertains the possibility that, having been born in Halicarnassus, Herodotus may have known Carian, he rejects this on the grounds that there is no evidence of such knowledge in his work. (2) Herodotus often uses language (along with other cultural practices such as clothing, food, government and religion) as a point of comparison between different groups of people. He makes use of two different Greek words--[TEXT NOT REPRODUCIBLE IN ASCII.] (meaning "tongue" and therefore "language") and [TEXT NOT REPRODUCIBLE IN ASCII.] (meaning "sound" and therefore also "language")--which he appears to use interchangeably, although the latter is also used of animal sounds (Harrison 1998:5). He compares the languages of different societies in three different ways. Sometimes he says that the language of one group is similar to, or resembles that of another; for example, the Caunians and the Carians (1.172.1), the Egyptians and the Colchians (2.105.1) and the Sagartians and the Persians (7.85.1). Elsewhere he describes the language of one group as different to that of another; for example, the Nasamonians and the Pygmies (2.32.6); the Scythians and the Amazons (4.111.1) and the eastern and non-eastern Ethiopians (7.70.1). Finally, he describes some languages as completely unique to one group of people; for example, that of the Argippaeans ([TEXT NOT REPRODUCIBLE IN ASCII.], 4.23.2), the Androphagi ([TEXT NOT REPRODUCIBLE IN ASCII.], 4.106.1) and the Trogodyte Ethiopians, who sound like bats ([TEXT NOT REPRODUCIBLE IN ASCII.], 4.183 .4). (3)
Secondly, Herodotus alludes to a number of sociolinguistic phenomena in his work. In the study of languages in societies, an important concept is language contact, something which occurs when people from different linguistic communities come into contact with one another (Mesthrie & Leap 2000:248). When languages come into contact with one another, they affect each other and change. Language shift, "the replacement of one language by another as the primary means of communication ... within a community" (Mesthrie & Leap 2000:253), occurs when a group of people who are exposed to a second language, start to use that second language, and over time, stop using the original language. Herodotus gives us a clear instance of language shift with respect to the Athenians who shifted from Pelasgian to Greek, ([TEXT NOT REPRODUCIBLE IN ASCII.], 1.57.3). He also describes language contact when he mentions that either the Caunians or Carians changed their speech to [TEXT NOT REPRODUCIBLE IN ASCII.] (approximate or approach) the other (1.172.1). On another occasion he says that the Phoenicians with Cadmus changed their language over time ([TEXT NOT REPRODUCIBLE IN ASCII.], 5.58.1). In contrast, he tells us that the Eritreans who were resettled by Darius maintained their original language ([TEXT NOT REPRODUCIBLE IN ASCII.], 6.119.4), showing resistance to language shift, a practice known as language maintenance (Mesthrie & Leap 2000:253), something possibly associated with an ethnic pride. From these examples, Herodotus has provided evidence (perhaps without fully understanding these things) that processes such as language contact, language shift, language death and language maintenance were as prevalent in antiquity as in modern times.
In addition to this, Herodotus provides evidence and even some discussion of dialects within a language. Most important here is his description of the four different Ionian dialects which he calls [TEXT NOT REPRODUCIBLE IN ASCII.], but also [TEXT NOT REPRODUCIBLE IN ASCII.]: those of Caria, Lydia, Samos, and a fourth shared by Chios and Erythrae (1.142.4). While Herodotus may have been monolingual, he must have been exposed to various dialects of Greek throughout his lifetime, especially those in Ionia, as is evident from his acknowledgement of their existence (cf. Robins 1999:11). (4) In ancient, as in modern times, it is not fully clear what constitutes a language as compared to a dialect (Harrison 1998:5). One useful, but not very accurate definition considers whether the variations of the language are "mutually intelligible", i.e. speakers of the one dialect should be able to understand speakers of another with only a little difficulty (Mesthrie 2000:9). It is not certain how the Greeks distinguished the two, but there clearly was a difference to Herodotus, because elsewhere, ignoring the dialects, he speaks of their "common language" ([TEXT NOT REPRODUCIBLE IN ASCII.], 8.144.2).
His portrayal of different languages and dialects within a language and his description of some languages as "similar" and others as "different" also show a sophisticated understanding of the links between languages and the existence of "language families". Harrison (1998:6) praises Herodotus for not lumping all non-Greek languages together. Herodotus recognises that some languages have characteristics in common with others, or are more like one language than another. In addition, he tries to apply what he knows about languages, and how they are related to one another, to formulate a hypothesis about what the original (now dead) "proto-" language of the Pelasgians was like. By observing the language spoken by their descendants in various places, he concludes that the Pelasgians spoke "a barbarous language" (i.e. a language which is not Greek): "[TEXT NOT REPRODUCIBLE IN ASCII.] " (1.57.1-3). In the same way that modern linguists try to trace related languages and hypothesise the features of the ancestors of those "language families" (e.g. Proto-Germanic, Proto-Indo-European or Proto-Bantu), so Herodotus acts a little like a historical linguist as he tries to find out what kind of language "Proto-Pelasgian" was.
In Herodotus' recognition of language families, we see something of his general portrayal of foreigners, which resulted in his being called [TEXT NOT REPRODUCIBLE IN ASCII.] (Plutarch, On the Malignity of Herodotus, 12). He did not divide the world into a binary "Greek" and the "Other" (barbarian), but was interested in learning about other cultures and finding similarities and differences between the different groups of people. While sometimes he does try to emphasise the difference between the Greeks and the Other; for example, his over-extended description of the Egyptians as "the complete opposite" ([TEXT NOT REPRODUCIBLE IN ASCII.], 2.35.2), at other times he is happy to portray different cultures as human and not that different from the Greeks; for example, the Lydians share many customs with the Greeks (1.94). This two-fold attitude carried over into his investigation of language. As there were relationships between different peoples, so were there relationships between their languages. Even among the Greek-speaking peoples, he saw that they were not all exactly the same (hence the various Ionic dialects). As there was contact and "cross-pollination" between different cultures (e.g. the Greeks learnt writing from the Phoenicians, 5.58, and about the gods from the Egyptians, 1.139), so was there cross-pollination, borrowing and language shift among their languages through language contact.
There are two cases in which Herodotus describes a language as being "halfway between" two other languages: these are the Geloni, who speak a language which is "part Scythian on the one hand, and part Greek on the other" ([TEXT NOT REPRODUCIBLE IN ASCII.] , 4.108.2), and the Ammonians, who speak a language "between" Egyptian and Ethiopian ([TEXT NOT REPRODUCIBLE IN ASCII.], 2.42.4). What exactly Herodotus meant by these descriptions is not clear. Perhaps these people were bilingual, and would frequently code-switch between the two languages so that it seemed that they were speaking a mixture of the two. Perhaps the two languages were distantly related and they spoke a dialect which was mid-way between them. Another tantalising possibility is that Herodotus is describing ancient pidgins or creoles. A Pidgin is a simplified mixture of languages used for communication between different groups (e.g. Fanakalo, used by mineworkers and on sugar plantations in South Africa). It lacks some of the formal features of proper languages and has no native speakers. Over time, a pidgin can develop into a full language which does have native speakers and becomes known as a "creole" (Pinker 1995:33). It is possible, though there may not be any real evidence, that the Geloni spoke a creole which had developed from a pidgin between Greek and Scythian. Since there was language contact in the ancient world, we should not be surprised to find evidence of pidgins and creoles as we do today.
There is another relevant case where Herodotus is quite likely describing some sort of creole. This is in his account of the Sauromatae who were descendants of Amazon women and Scythian men (4.111-117). The women tried to learn the Scythian language, but did so "imperfectly" ([TEXT NOT REPRODUCIBLE IN ASCII.]), and so their children spoke an "incorrect" version of Scythian ([TEXT NOT REPRODUCIBLE IN ASCII.] (5) , 4.117.1). It seems that he is describing the equivalent of modern creoles (e.g. those of the Caribbean) or "New Englishes". (6) Both these types of language sound "ungrammatical" to speakers of the standard dialect, perhaps in the same way that the Sauromatae dialect sounded "incorrect" to the Scythians. Whether Herodotus is describing creoles, the equivalent of "New Englishes" or simply instances of bilingualism, he has described the results of contact between languages with detail and accuracy.
The next aspect of language which Herodotus deals with is that of foreign or second language learning. Harrison (1998:3) criticises Herodotus for only mentioning "one clear instance of a Greek speaking a foreign language" and "a single instance of a Persian speaking Greek". But in Herodotus' defence, he does mention a number of examples of people learning or being taught another language, often Greek. The first example is of Cyarxes, an early king of Media who took in some Scythian nomads. He entrusted some Median boys to their care whom they were to teach "their language ([TEXT NOT REPRODUCIBLE IN ASCII.]) and the craft of using the bow ([TEXT NOT REPRODUCIBLE IN ASCII.])" (1.73.3). The second example is similar; Psammetichus of Egypt commanded some Ionian and Carian men to look after some Egyptian children and teach them the Greek language ([TEXT NOT REPRODUCIBLE IN ASCII.], 2.154.4). These children became the class of Egyptian interpreters ([TEXT NOT REPRODUCIBLE IN ASCII.]), and were an important part of Egyptian society, used in communications between Greece and Egypt (2.154.4). Another story that he gives is of Scylas, king of Scythia (4.78). His mother was a Greek woman from Istria and so Scylas grew up learning the Greek language and alphabet ([TEXT NOT REPRODUCIBLE IN ASCII.]). This proved to be his downfall, as the Scythians despised the Greek customs that he followed; especially his initiation into the cult of Dionysus (which even the gods disapproved of) and he was eventually beheaded by his own brother (4.78-80). Here we see early evidence of someone suffering discrimination for not sticking to their own culture and language. This is an important issue in modern sociolinguistics (for example the choice of English versus Native African Languages in South Africa).
There is another example of second language learning when some young Pelasgian men from Lemnos abducted some Athenian women and had children by them (6.138). These children were brought up according to the Attic language and manners ([TEXT NOT REPRODUCIBLE IN ASCII.], 6.138.2). But they became haughty and proud, lording it over their Pelasgian comrades ([TEXT NOT REPRODUCIBLE IN ASCII.] , 6.138.2). In this story we see something more of the identity and pride that can be associated with language. This is another phenomenon which is important in modern sociolinguistic study, and we again have evidence of it in modern day South Africa, where the prestige associated with English results in linguistic discrimination.
The final example of second language learning is that by the Amazon women (already discussed) from whom the Sauromatae were descended. In this case, it was adults, not children who learnt the language, and they did so imperfectly. This is significant for modern language acquisition theory which claims that children are born with an innate blueprint for acquiring languages (called Universal Grammar or The Innateness Hypothesis). (7) Many believe that this blueprint disappears (or is inactivated) by adulthood and accounts for why it is more difficult for adults to learn a second language (Pinker 1995:290). Again, Herodotus has provided believable evidence about language, without necessarily realising or understanding it.
As Harrison (1998:1) points out, "Herodotus was not ... a professional philologist [and] we have no reason ... to expect a consistent or rationalized theory of the nature of language Most of what I have described so far are things which Herodotus has subconsciously revealed in his general inquiries and descriptions of societies. He does not have a lot of what is called "meta-linguistic knowledge" (i.e. a conscious awareness of the workings of language). But there is one instance in which he seems to hint at an even deeper knowledge of language and its nature. There is much debate among linguists on whether language is a result of nature or nurture. We get a slight glimpse into Herodotus' attitude towards this in his story of Psammetichus and his experiment to determine the first race on earth. Herodotus (or at least Psammetichus whom he does not criticise) must have believed that language was somehow innate because it was expected that even without any verbal input, children would eventually use language. But at the same time, he recognised, that children learned to speak whichever language they heard around them (hence the command that "there never be any sound in their presence", [TEXT NOT REPRODUCIBLE IN ASCII.] , 2.2.2). This resembles the modern view that while the capacity to acquire language is innate, input from a specific language is necessary to acquire that specific language. Here he deviates from the modern view with the additional assumption that some languages are older than others, their age corresponding to the age of the societies which spoke them. Herodotus (or Psammetichus) believed that without any input, the children would revert to whatever was the oldest language. This contradicts the modern view that language (or at least naming of words / vocabulary) is arbitrary in all languages of the world.
The final area related to language which I will discuss deals with the issues of translation and interpretation. As mentioned, Herodotus is often criticised for ignoring this issue (Harrison 1998:3-4, Waters 1985:79 and Robinson 2002:1). While it is true that Herodotus does not often mention the presence of interpreters either in conversations which he describes between people who speak different languages, or between the people who told him stories and himself, he does refer to interpreters a number of times in his work. The word he uses for "interpreter" is [TEXT NOT REPRODUCIBLE IN ASCII.], from Hermes, the messenger of the gods. Firstly (2.154), he tells us of the class of Egyptian interpreters (which will be discussed again later). He also mentions that when the Scythians travelled through the remotest parts of the Black Sea region, they made use of seven interpreters who knew seven languages ([TEXT NOT REPRODUCIBLE IN ASCII.] , 4.24). There is also the story about Cambyses who performed an experiment with a group of Greeks and one of Indians to compare their funeral practices (3.38). Herodotus tells us that this conversation was "conducted through the aid of an interpreter" ([TEXT NOT REPRODUCIBLE IN ASCII.], 3.38.4). We also have the well known story of Cyrus and Croesus. The important conversation between them (after Cyrus decides not to kill Croesus, and when Croesus tells him to beware of his fortunes as he too might lose them), is conducted through interpreters (1.86). There is yet another account in which Sylosson, a Sammian prince, speaks to Darius through the aid of interpreters (3.140).
Herodotus also shows recognition of the fact that language can be a barrier between different cultures when he describes the first encounter between a Scythian man and an Amazon woman. He tells us that "they used hand signals to communicate" ([TEXT NOT REPRODUCIBLE IN ASCII.], 4.113.2). From these passages we can see that Herodotus was not ignorant or careless when it came to issues of interpretation. It would have admittedly been tedious and unnecessarily repetitive if he had mentioned interpreters every single time they were present. So instead he simply mentions them here and there, to remind us that there was inter-lingual communication taking place. In this he is no different to (in fact perhaps more accurate than) modern writers (of fiction and non-fiction) who often neglect to refer to and deal with the problem of language barriers.
In terms of Herodotus' personal encounters with interpreters, we have even less evidence. The only overt reference which I have found is when he speaks of an Egyptian interpreter of hieroglyphics ([TEXT NOT REPRODUCIBLE IN ASCII.] , 2.125.6). Despite the criticism levelled against Herodotus for omitting to discuss his interpreters (cf. Chamberlain 1999:265), I would argue that Herodotus does provide ample evidence for (at least) his Egyptian interpreters. I find this in the passage already alluded to, in which he tells us about the "class of Egyptian interpreters" (2.154). There are a number of important things to note about these men. Waters complains that:
Effective rendering of one language into another requires an expert in both languages; many Egyptians ... may have had a superficial knowledge of Greek, but that would be quite inadequate for the answering of many of Herodotus' questions (Waters 1985:79).
But Herodotus' account of the Egyptian interpreters paints a very different picture. He tells us that these Egyptians were taught Greek from childhood, which means that they would have acquired near-first language proficiency. In addition, they were originally taught by native Greek speakers, and so their Greek would have been almost as good as any native speaker of Greek. We would assume that since these boys grew up in Egypt, they would have had an equal knowledge of the Egyptian language and have been native bilinguals. If Herodotus' account is accurate, then the problem of finding adequate interpreters for his Egyptian inquiries is non-existent. There was a group of men who had been brought up for the very purpose of facilitating Greek and Egyptian communication. Herodotus emphasises this when he says that:
Since these [people] settled in Egypt, we Greeks, through interacting with them, know thus, with certainty, all the things which happened concerning Egypt, starting from the reign of Psammetichus and those [who came] after (2.154.4).
With these words, Herodotus is indirectly saying that "it was through the translation of these people that I accurately learnt what I know about Egyptian history". It would seem that Herodotus is not simply making up this story because he does not refer to just a group of translators, but elsewhere he speaks of a whole class of interpreters, one of seven classes in Egypt ([TEXT NOT REPRODUCIBLE IN ASCII.], 2.164). They must have been a class of significant size and status to have been mentioned as one of the Egyptian social classes; along with the priests, warriors, herders, tradesmen and boatmen.
Contrary to those who claim that Herodotus ignores the problem of his interpreters, it seems that he gives us a detailed explanation of what happened in Egypt. In Egypt, there was a class of men, who were brought up from childhood to be fluent in Greek. It was the occupation of these men to act as interpreters, and so when someone like Herodotus came to Egypt wanting to learn about their culture, he would naturally have approached them to act as interpreters. Because this was their occupation, and they had been trained from childhood by native speakers, their reliability would be unquestioned, and therefore there was no further need for Herodotus to discuss them, nor to doubt the accuracy with which they translated for him.
Language may never have been Herodotus' main focus, but it is clear that he did deal with it throughout his Histories. It has been seen that he used language as a criterion for comparing cultures. He also revealed that a number of sociolinguistic phenomena were present in the ancient world, by alluding to concepts such as language contact, shift, death and maintenance. He revealed a sophisticated understanding of the relationships between different languages and referred to dialects. He also described examples of what are either bilingual communities, or else people who made use of the ancient versions of "New Englishes" or pidgins and creoles. In addition to this, Herodotus provides examples of second language learning in the ancient world by both children and adults, and gives us a hint about what he believed about the nature of language. Finally, he gives us a number of examples of translators, and a detailed account of the origins (and therefore existence) of a class of efficiently trained interpreters in Egypt. The nature of Herodotus' inquiry determined that he could not avoid dealing with different languages. He seems to have had a fairly in-depth understanding of issues around language and multilingualism in the ancient world, and there is much that we can learn from him in this regard.
Chamberlain, D 1999. On atomics and metarrythmic translations in Herodotus. Arethusa 32:263-312.
De Selincourt, A (trans.) 1972. Herodotus: The Histories. Middlesex, England: Penguin Books.
Harrison, T 1998. Herodotus' conception of foreign languages. Histos 2. Accessed from: www.dur.ac.uk/Classics/histos/1998/harrison.html.
Horrocks, G 1997. Greek: A history of the language and its speakers. Harlow, England: Pearson Education Ltd.
Kaschula, R H 2007. Communicating across cultures in South Africa: Speaking in tongues. Reprinted in Introduction to language study. 151-158. Durban: University of KwaZulu-Natal, Linguistics Department.
Legrand, P E (ed.) 1960-1973. Herodote: Histoires. Paris: Les Belles Lettres.
Mesthrie, R & Leap, W L 2000. Language contact 1: Maintenance, shift and death. In Mesthrie, R, Swann, J, Deumert, A & Leap, W L (eds.), Introducing Sociolinguistics, 248-278. Edinburgh: Edinburgh University Press.
Mesthrie, R 2000. Clearing the ground: Basic issues, concepts, and approaches. In Mesthrie, R, Swann, J, Deumert, A & Leap, W L (eds.), Introducing Sociolinguistics, 1-43. Edinburgh: Edinburgh University Press.
Pinker, S 1995 (1994). The language instinct. London: Penguin Books.
Rawlinson, G (trans.) 1910. Herodotus: The History of Herodotus. London; Toronto; New York: Dent, Dutton (Everyman's Library).
Robins, R H 1967. Greece. In A Short History of Linguistics, 9-44. London; Harlow: Longmans, Green & Co. Ltd.
Robinson, D 2002 (1997). Herodotus. In Western translation theory: from Herodotus to Nietzsche, 1-4. Second Edition. Brooklands, Manchester, UK: St. Jerome Publishing.
Waters, K H 1985. Sources of information. In Herodotus the historian: his problems, methods and originality, 76-95. London; Sydney: Croom Helm.
The paper judged to be the best student essay submitted to Akroterion by December 8, preceding publication of the volume for that specific year, is published annually as the CASA / KVSA Essay. The competition, which is sponsored by the Classical Association of South Africa, is open to undergraduate students every year and to Honours students in even-numbered years. The winner receives a cash prize of R800.
Sonja Gammage, Greek 3B (University of KwaZulu-Natal)
(1) There has been some investigation of the issue of language in Herodotus. For example, David Chamberlain 1999 defends and discusses Herodotus' "linguistic competence", but he focuses mainly on Herodotus' interpretation of names from foreign languages, specifically those of the Persian kings and the Egyptian gods. Thomas Harrison 1998 gives a fuller account of "Herodotus' conception of foreign languages", but again comes back to issues of interpreting specific foreign terms. I attempt to give a more general account of what we can learn about the sociolinguistic situation and multi- (or inter-) lingual societies of Herodotus' day.
(2) The Carians dwelt in a region of south-western Asia Minor, which included the Greek city of Halicarnassus, where Herodotus was born (OCD3 s.v. Caria; Herodotus I). Carian was a nonGreek language, possibly from the Anatolian language family, which is an Indo-European group that includes Hittite and Lydian (OCD3 s.v. Anatolian languages).
(3) All translations are my own. I did, however, consult the translations by de Selincourt 1954 and Rawlinson 1910. The Greek text used was that of Legrand 1960-1973.
(4) In addition to this, he may have been exposed to the Doric dialect since Halicarnassus was originally Dorian, and had only recently been "Ionicised" in Herodotus' day (Horrocks 1997:23).
(5) The word aoXoiKiZovxsq, meaning "to speak incorrectly"--the origin of our English word "solecism"--is named after the Soloi of Cilicia who spoke a corrupt dialect of Attic (cf. LSJ9 s.v. aoXoiKoq II).
(6) The term "New English" refers to a dialect of English, which originates as the second language of a community. Because it is not their mother tongue, speakers acquire it imperfectly, and it contains features of the substrate language. Over time, language shift occurs, and it becomes the mother tongue of the community. The resulting dialect contains a number of features which are ungrammatical in the standard dialect but have been carried over from the original language. Such dialects are called "New Englishes" because most of the modern examples which have been studied are new dialects of English resulting from the days of the British Empire (Mesthrie & Leap 2000:310). Examples include: Black American English and South African Indian English. Herodotus may be describing an ancient equivalent of such a new dialect.
(7) Terms made popular in linguistic theory through the influence of Noam Chomsky.