«Kelly R. Bristol University of South Florida Phi Alpha Theta Regional Conference April 14, 2012 Slavery was a “necessary evil” in ancient Greece ...»
You Say Dmos, I Say Doulos:
Language and the Development of Greek Chattel Slavery
Kelly R. Bristol
University of South Florida
Phi Alpha Theta Regional Conference
April 14, 2012
Slavery was a “necessary evil” in ancient Greece and the ancient world as a whole.1 The
agrarian-based societies of antiquity needed massive amounts of manpower to function, but also
had to have stable economies. To achieve this, it was commonplace for these societies to resort to treating certain humans like commodities; they could be bought, sold, trained, and controlled, all while providing little to no monetary compensation for the slaves’ work.2 Physical, psychological, and fiscal control of slaves allowed for the consistent production of affordable goods, which were necessary for the society’s survival. In essence, slavery was the backbone of antiquity, and exposing it would lead to a deeper understanding of the ancient world.
Unfortunately, the extant source materials we have for slavery—much like other areas of scholarship on the ancient Greece—are scarce and are problematic at best. There is a dearth of archaeological evidence which can be directly related to Greek slaves, and very little written material which mentions slavery in a meaningful capacity. Fully understanding the slave condition in Athenian society is not possible, but we may be able to gain additional insight by using what evidence we have in different ways.
The most important building block of any given society is its language. Language provides a sense of unity and is the mode of transmission for ideas and knowledge. The language of a society shapes and is shaped by those who use it, becoming a historical tool itself. The language of the ancient Greeks is a particularly rich source for study, as its dozens of dialects eventually coalesced into a language that was known during the Hellenistic period (323-31 BCE) as koine, or common, Greek. What this means for historians is that the extant texts from ancient Joseph Vogt, Ancient Slavery and the Ideal of Man, trans. by Thomas Wiedemann (Cambridge, MA: Harvard University Press, 1975), 25. For the purposes of this paper, the use of the word “Greece” or “Greek” refers specifically to Athens, with exceptions noted as necessary.
For a sociological perspective on matters of power, property, and slavery see Orlando Patterson, Slavery and Social Death: A Comparative Study (Cambridge, MA: Harvard University Press, 1982), 17-28.
Greece preserve that living, changing language, and this can be used in a variety of ways. I will be using language to examine shifts in attitudes toward slavery from the Archaic period through the Classical period, roughly spanning the early 8th through late 4th centuries BCE.3 More specifically, I will explore the distinct change in the commonly-used word for slave—from dmos to doulos—that seems to coincide with the emergence of the ‘true’ chattel slave in Athens. This change in language and attitude toward slavery played an understated yet critical role in the major societal changes that occurred during the Archaic and Classical periods of Athens.
In order to discuss this topic, it is necessary for me to discuss the history of the language, as well as the known etymology of dmos to doulos. Additionally, I will also discuss the uses of these words in context. For the general study of the Greek language, I have attempted to trace the lineage of the language and of key words and the concepts associated with them. The primary sources that I have consulted most frequently are: Homer, Hesiod, Aristotle, Hippocrates, Plato, and Plutarch. These authors reflect the popular vernacular of their respective times, with the notable exception of Plutarch, who was a 1st c. CE biographer. Plutarch’s works, though written several centuries after the period I am discussing today, nevertheless remain valid as they are compiled from older authors of Greek and Roman history. The language and vocabulary used in his Greek Lives would likely have been influenced by that of the original non-extant sources, especially if there were words for which Plutarch had no modern equivalent. This also holds true for the pseudepigrapha written by the anonymous Pseudo-Plutarch.4 Because pseudepigrapha are Tracking language changes after the reign of Alexander the Great (r.336-323 BCE) becomes difficult due to the spread of koine Greek.
Michael J. Edwards, “Notes on Pseudo-Plutarch's Life of Antiphon,” The Classical Quarterly (new series) 48, no.
1 (1998): 82, http://www.jstor.org/stable/639753 (accessed November 10, 2011). Edwards is just one of a multitude of scholars who calls the veracity of Plutarch’s Vitae Decem Oratorum (part of the Moralia) into question. Because the source that I have used for this paper still refers to the author of this work as Plutarch despite its questionable nature, I will do so as well in an effort to avoid further confusion.
based upon emulation, Pseudo-Plutarch would also have sought to reflect the same language influences as the authentic Plutarch. The language used by these authors is crucial in understanding the changes in vocabulary that occurred. Thus, we must briefly examine the Greek language itself for clues which might help explain some of the changes in the language of slavery.
The Greek language as we know it today is the result of over 3,000 years of development.
It is one of the oldest living languages in the world. The roots for the Greek language are considered to be Indo-European, with the earliest examples of what would become Greek being the languages known as Linear A and B.5 Linear A has yet to be deciphered, but Linear B was cracked by Michael Ventris and John Chadwick in the 1950s.6 Linear B is a combination of ideograms—pictures that represent ideas and concepts—and a syllabic script. Though some disputes continue over specific words, the word for slave (do-er-o, feminine do-er-a) is commonly held to be an accurate translation.7 This word is similar to the typically post-Homeric term doulos, and despite the loss of writing during Dark Age Greece (c.1200-800 BCE) we can make the correlation between the two terms.8 What remains unclear is when and where the Mycenaean do-er-o is lost and the Homeric term dmos comes into use.
For much of antiquity, Greece was nothing more than a geographic area consisting of many scattered settlements. These settlements had been colonized by a multitude of peoples from around the Mediterranean, who imposed their own variations on the emerging spoken Greek Francisco Rodriguez Adrados, A History of the Greek Language: From Its Origins to the Present, translated by Francisca Rojas del Canto (Leiden: Brill, 2005), 3-7; Geoffrey Horrocks, Greek: A History of the Language and its Speakers (New York: Addison Wesley Longman, 1997), 11.
Horrocks, Greek: A History of the Greek Language, 3.
John Chadwick and Lydia Baumbach, “The Mycenaean Greek Vocabulary,” Glotta 41 (1963): 187, http://www.jstor.org/stable/40265918 (accessed November 9, 2011).
Yvon Garlan, Slavery in Ancient Greece, translated by Janet Lloyd (Ithaca: Cornell University Press, 1988), 25-29.
language. The major dialects included Ionian, Dorian, and Aeolian.9 Not unlike dialects of a modern language, many of the differences in the Greek dialects involved pronunciation and vocabulary. Following the Greek Dark Ages, writing re-emerged with a new alphabet that is believed to be derived from the Phoenicians.10 This Archaic period also heralded the development of the polis, or city-state, which had coalesced, in part, around areas where the speakers of these Dark Age dialects had concentrated themselves. As each city-state adopted the alphabet these spoken differences began to manifest themselves in the written form. Modern linguists attest that there are at least fourteen known separate alphabet systems during this formational period.11 The lack of a standardized spelling system created further disparities within the early written Greek language. During the early Archaic period, when the works of Homer and Hesiod were beginning to cross over from oral tradition to written literature, the Greek alphabet was still in flux—letters were being shifted around as some dialects became more prominent than others. The pure Ionian dialect has its distinctions, which we see in Herodotus.
Lyric poets typically employed the Doric dialect, and the Aeolian dialect is preserved in fragments of Sappho.12 In addition to the alphabetic differences, the vocabulary problems posed by these multiple dialects are numerous. As the use of these various dialects fell out of fashion, it became increasingly more difficult for transcribers to accurately copy texts. Ancient scribes did not have dictionaries like we do today and had to compensate accordingly. Words were substituted for those that were the closest in pronunciation or spelling, and sometimes omitted completely. The same held true for differences between the alphabets, with letters like the For comprehensive discussions of the evolution of Greek dialects, see Adrados, A History of the Greek Language, 59-89, and Horrocks, Greek: A History of the Language, 6-15.
Adrados, A History of the Greek Language, 64.
Adrados, A History of the Greek Language, 64-66.
Adrados, A History of the Greek Language, 106-109.
digamma (Ϝ) commonly being left out of copied texts. It was not until the 6th c. BCE that an effort was made to create a standardized written form of Greek to be used for official purposes.
This ‘official’ form was derived from the Attic tongue of the Ionian dialect and is what scholars refer now to as Classical Greek.13 Despite this ‘official’ written form, other dialects’ alphabets and spelling persisted, but overall a greater level of consistency in vocabulary emerged.
Problems with the written record also arise in dictation. Just as you or I might make an honest mistake due to misheard speech, so could ancient scribes. We can never really measure the extent of the impact that these unknown or misinterpreted words had in our ancient texts, but this should give ancient historians and philologists cause for concern. In some cases the language discrepancies can call into question the reliability of the ancient source while in others, language consistency can help to legitimize the source. A prime example of how the veracity of a work can be questioned is with the linguistic analysis of the works of Homer.14 The main hallmark of the Archaic period is the re-emergence of writing and, in particular, the birth of Western literature with the Iliad and Odyssey of Homer. The Homer that most of us are familiar with is believed to have lived around the mid 9th c. BCE, though modern scholars have raised what is often referred to as the “Homeric Question”.15 The “Homeric Question” challenges not only the authorship of the works attributed to Homer, but even his very existence, with linguistic cues forming the bulk of scholarly evidence. Regardless of the questions surrounding his existence, Homer is often paired with another poet, Hesiod, who lived around Horrocks, Greek: A History of the Language, 17-23.
Gregory Nagy, Homer’s Text and Language, (Urbana, Ill: University of Illinois Press, 2004), 25-39. This work is an examination of the various problems with the text and their roots.
the 8th c. BCE.16 However, neither poet is believed to have written his own works. There are some scholars who claim that the works of Homer were committed to writing possibly as late as during the reign of Peisistratus, around the mid-6th c. BCE.17 Because the Ionic-Attic dialect had not yet become the prevailing written form of the Greek poleis, the word forms we see in Homer and Hesiod are largely referred to as Homeric. Some authors argue for the Homeric dialect being pure Ionic-Attic, but further investigation reveals too many differences between the Homeric and Ionic-Attic standard forms.18 Because the language used by Homer and Hesiod (and others from the period who come to us in fragmentary form) is consistent, we can not only firmly place these works within the same time period but also see patterns emerge from the texts. For example, most modern beginning students of Classical Greek are taught that the word doulos means slave, with little to no indication that there are any other terms associated with slaves. Not so! In fact, the word doulos scarcely appears in either Homer or Hesiod. What does appear as a generic term for slave is the word dmos. Conversely, dmos hardly appears after Homer and Hesiod, and is especially rare after the turn of the 5th c. BCE.19 If the concept of what a slave was in Athenian society had remained consistent over time, we would expect to see some type of continuity between the words. We do not. Etymologically, these words have no relation to each other. Something had changed in the very nature of what a slave was and what he or she meant to free Greeks.
Ibid.; Martin Litchfield West, “Hesiod,” in The Oxford Classical Dictionary, 3rd edition, ed. Simon Hornblower and Anthony Spawforth (Oxford: Oxford University Press, 2003), 700.
Nagy, Homer’s Text and Language, 10-11.
Adrados, A History of the Greek Language, 142-160; for a contrasting view, see G. P. Goold, “Homer and the Alphabet,” Transactions and Proceedings of the American Philological Association 91 (1960): 272-291, http://www.jstor.org/stable/283857 (accessed November 18, 2011).
See Appendix. This listing is by no means complete, but is intended to illustrate general patterns in the primary forms of the words for slave. For example, δοῦλος appears once each in the Iliad and the Odyssey, but nowhere in Hesiod. Compare to δμώς and its derivatives.