Eleanor Dickey, Greek Forms of Address. From Herodotus to Lucian. Oxford: Oxford University Press, 1996. Pp. 336. $75.00. ISBN 0-19-815054-7.
Reviewed by Helma Dik, Department of Classics, University of Chicago.1
Word Count: 3,618.'Would it be of any use, now,' thought Alice, 'to speak to this mouse? Everything is so out-of-the-way down here, that I should think very likely it can talk: at any rate, there's no harm in trying.' So she began: 'O mouse, do you know the way out of this pool? I am very tired of swimming about here, O Mouse!' (Alice thought this must be the right way of speaking to a mouse: she had never done such a thing before, but she remembered having seen, in her brother's Latin Grammar, 'A mouse-of a mouse-to a mouse-a mouse-O mouse!')2
Eleanor Dickey's study, Greek Forms of Address, is the first full-scale study of the ancient Greek address system as it can be reconstructed from the most important surviving documents: Greek prose literature. Thanks to D.'s decision to include Lucian in her study, we can be fairly confident that a Greek Alice would indeed have used 'O mouse' to address a mouse (section 3.10, p. 185: "In our data animals are always addressed with terms indicating their species").
D. is not the first to study the Greek address system, or what are called Anredeformen in the existing literature.3 However, in the past scholars concentrated on usage in poetry and various points of detail, such as the nominative-for-vocative construction and the vexed problem of when W)= is used and when it isn't. D. does oblige philologists with an abiding interest in W)=, but she has much more to offer in her book. Her goals are wide-ranging and more fundamental than those of her predecessors, which inevitably leads to some observations that are less than surprising; on the other hand, she shows convincingly where many of these earlier studies have gone wrong; more importantly, she has many interesting observations of her own, on Greek forms of address and their development through time, on usage in individual authors, and last but not least on the interpretation of specific passages (conveniently listed in an index locorum) in the literature.
D. sets out to present the Greek address system in the context of current theory in the field of sociolinguistics, and takes pains to adduce parallels from languages as familiar as English and as far apart as Kannada and Kazakh. Thus we hear that the address 'husband' was acceptable in Elizabethan English and that cooks are commonly addressed with maharaj 'emperor' in Hindi. D.'s bad fortune is, as many readers will realize, that there is not an awful lot of variation in the Greek address system. FN ('first name'; see below) will do in the vast majority of situations that we encounter in texts: adult male citizens addressing adult male citizens.4 But this is not the whole story, even if we ignore the occasional entertaining aside about terms of endearment in Icelandic (p. 166).
In the introductory first chapter, D. defines her subject and introduces some important distinctions. D. restricts her study to so-called 'free' forms of address, so in 'Mary, could you please open the window?' the use of 'Mary', not 'you' is at issue. For Greek this means in practice a restriction to vocatives mostly. Section 1.2 introduces some of the findings from sociolinguistics and the variables that will play a role in the description of address in any language. A system of rules for forms of address has to take into account the relationship between speaker and addressee. It is not enough to correlate merely the identity of the addressee and the form of address used and the social context (the setting, the audience, the topic of discourse) of the utterance. The scholar of address has to consider that there may be different systems within one language, in use by different groups in a society. The choice of register and politeness phenomena can also be important factors in determining address use. An important distinction is further to be made between referential and address usage of words. In the passage from Alice the mouse (reference) was addressed with 'O Mouse' (address usage). In most cases there will be a common 'ancestor' for referential and address use but the two can grow apart considerably. We will come across examples of this below, such as the difference between referential and address usage of basileus 'king'.
Section 1.3 gives an overview of previous work on Greek addresses and 1.4 discusses the material used for the study: addresses in prose from Herodotus to Lucian, with frequent excursions to usage in poetry for purposes of comparison, in particular Aristophanes and Menander. D.'s corpus comprises 11,891 prose addresses plus 1,693 instances collected from Aristophanes and Menander. More interestingly, the corpus consists of over 350 different forms of address, not counting, of course, different names and ethnics. Matters to do with W)= are relegated to a separate section and use or non-use of W)= is not taken into account in the main discussion of forms of address.
Chapter 2 brings up what seems to be an obligatory exercise in many books introducing theory in classics, and reminds me of a lot of introductory chapters from the past three decades, at least. Do classical linguists applying modern linguistic theory still have to give an apologia pro vita sua? I should think not, but perhaps this is due to my own prejudice in favor of such approaches. At any rate, for those stubborn readers who remain unconvinced, here again is the argument, specifically this time for the application of sociolinguistics to Greek literature. The conclusion to this chapter has been given a 'cliff-hanger' presentation which I personally don't like ("At this point we cannot say how well these procedures will allow us to reconstruct conversational Greek; we must first proceed with the reconstruction and return to this issue in the Conclusion"). Surely D. had finished her research when she wrote this. She knew all right. I, the reader, just have to wait until she thinks it is time for me to hear her answer. I, for one, get irritated when these cliff-hangers appear too often, especially without reference to where this particular thread will be picked up again.
Chapter 3 forms the centerpiece of this book, with over 140 pages. All forms of address are discussed, starting with the basic form of address, 'FN'. Using FN, the normal term in address studies for use of the first name, is a bit unfortunate for Greek where, after all, there was no such thing as a 'last name', but the meaning is clear enough. Address by given name was the default option when the speaker knew the addressee's name, and well over sixty percent of singular addresses are by FN alone. The exceptions to this general rule vary by author and genre. Deviation from FN between males of roughly equal status suggests a difference in status and can thus be used as display of affection, subservience or, conversely, in insults. Women are rarely addressed by name until later periods (Lucian). Patronymics are rare in D.'s corpus, and that naturally makes it more difficult to come to conclusions about their use. D. concludes that outside of Plato (where practically all patronymics are put in Socrates' mouth) patronymics are deliberately avoided in neutral statements. They can be found in especially formal, deferential or courteous speech on the one hand and in very negative statements on the other hand. D. says Homeric evidence (Il. 10.68-69, PATRO/QEN E)K GENEH=S O)NOMA/ZWN A)/NDRA E(/KASTON, PA/NTAS KUDAI/NWN) implies that patronymics honoured the addressee. D. does not cite a prose instance, Thuc. 7.69.2, where Nikias "AU)=QI TW=N TRIHRA/RXWN E(/NA E(/KASTON A)NEKA/LEI, PATRO/QEN TE E)PONOMA/ZWN KAI\ AU)TOU\S O)NOMASTI\ KAI\ FULH/N," a speech in a pretty desperate situation for the Athenians, so definitely not a neutral statement. However, neither does this situation fit D.'s description of deferentiality, let alone hostility. The occasion is an important one, and calls for an 'epic' speech. As Hornblower suggests (Greek Historiography, p. 9), "given in full, this speech might have looked distinctly Homeric." Despite the enormous amount of material D. has worked through, there will always be suggestions for additions and here is one: besides searching all texts for vocatives, a search for words like O)NOMA/ZW and cognates, or a rare word like PATRO/QEN, could have yielded valuable additional evidence, especially in cases where actual usage in the texts is rare, as is the case with the patronymics.
Kinship and age terms (ch. 3.2) are an especially pernicious area in address studies. D. rightly points out that words in this category can in address usage deviate greatly from their referential meaning. One need only think of (Am.) English 'brother' to realize the problem. There has been much speculation by D.'s predecessors and commentators about the terms PAI= and TE/KNON but D. concludes sensibly that the vocative TE/KNON is "purely and emphatically a kinship term, while PAI= can indicate both youth and kinship". This is borne out by the fact that speakers who are not related to addressees tend to use PAI=, not TE/KNON, and when they do use TE/KNON, they will usually have a special, parent-like, relationship to the addressee. TE/KNON furthermore is preferred by parents when the kinship relation is being emphasized, for instance in emotional scenes. Because of the preponderance of the kinship, not age, meaning in TE/KNON, it can be used to address adults, where PAI= is restricted to children and youths for speakers other than the addressee's parents. D. argues (70) that PAI= as used to slaves originates in its use as an age term rather than as a kinship term which also 'explains' why the pure kinship term TE/KNON is not so used.
Given the Greek fascination with young males, there was enough material for D. to devote a separate section to addresses for young men. Interestingly, there is again a discrepancy between reference and address. Young men can be referred to as MEIRA/KION or NEANI/AS with PAI=S being diminutive, but PAI= can be used to address them without this derogatory connotation, and in fact it is the age term used to address Theaetetus in the Theaetetus, Sophist and Politicus. D. does not speculate on the reason why Protagoras uses NEANI/SKE when delivering a sales pitch to a possible pupil (Prot. 318a6) and why Socrates uses it to Phaedrus (at Phdr. 257c8). But since both addresses are also used in combination (Leg. 904e5) perhaps it is indeed wiser not to speculate. In general, it seems to be true that PAI= used to young men and adults is affectionate, whereas NEANI/A used to adults is insulting.
This last observation brings me to another group of age terms with potential for insult: GRAU= and GE/RON. These forms of address are far less frequent in prose and in Menander than they are in Aristophanes, which leads D. to conclude that the frequency in actual usage will have been lower than in Aristophanes, but it remains unclear whether these terms are to be considered insults or neutral expressions, as seems to be the case in tragedy.
Moving on to A)/NER and GU/NAI, it is not exactly surprising that D. finds that husbands do indeed use 'wife' to address their spouses, but that 'husband' is not a frequent form of address. After all, adult males will not have been identified as 'husband of' but as individuals in their own right, unlike women. D. dispels the myth that GU/NAI is a term of respect or affection. To the contrary: the evidence suggests that use of 'FN' is preferred in more emotional, affective contexts ( pp. 86; 224). The use of titles is rare in Greek addresses, but D. unearths some interesting material regarding the use of BASILEU= and DE/SPOTA to monarchs. Where Spartan, Roman and other 'civilized' kings are regularly referred to as 'BASILEU/S', they are hardly ever addressed by title. Instead they will be addressed by name, and the vocative BASILEU= is mostly used to address absolute monarchs. The difference with DE/SPOTA lies in the status of the speaker: when DE/SPOTA is used, the speaker counts himself among the addressee's subjects or slaves, and this need not be the case with BASILEU=.
This review would go on far too long if it were to cover exhaustively what D. does cover exhaustively in this chapter. I will skip, for instance, D.'s treatment of insults (3.7.1 'Classical insults'; 3.7.2 'Post-classical insults'), in hopes that readers of this review will be tempted to look at D. themselves, and I will here simply voice my hope that D. will in future fill the regrettable lacuna she herself observes. We need (that is, I would love to see) a full study of Greek insults.
I will, however, say a few words about D.'s treatment of friendship terms in Plato, because this is one of those cases where she goes beyond present scholarship and shows how an approach like hers will come up with results that scholars working on individual texts ignore at their peril. The question at hand is the following. It has been maintained that Socrates' use of FI/LE and similar terms expresses a peculiar kind of philosophical companionship. Careful analysis leads D. to conclude that what Socrates in fact does in using this form of address is show superiority, not affection. This is particularly obvious from the fact that even Meletus in the Apology is "dear Meletus". Plato has Socrates say "dear X" when he is dominating the argument. Since, as D. shows, the same use of 'dear' turns up in Xenophon's portrayal of Socrates, but is rare in the rest of the corpus, we may tentatively conclude that we are dealing with an idiosyncrasy of the historical Socrates, which fits in well with the picture of an accomplished ironist. D. does not explicitly pronounce on whether Socrates' deviant use of patronymics (above) can be explained in the same way as that of friendship terms; but apparently there are no such indications in the context.
A final section that I would like to single out, especially for the benefit of those of us who dream of one day taking a time machine to fifth-century Athens, deals with indefinite addresses. How does one say "Hey mister!" in correct Attic? D. provides us with the answers in section 3.5. Being from out of town, you can expect to be addressed with XE/NE. However, once people know you personally, this address can indicate their irritation and you should take it as an insult. You yourself had better not address Athenians with XE/NE -- after all, it is you who are the stranger. The appropriate expression you should use to a stranger is A)/NQRWPE. As pointed out above for GU/NAI, conventional wisdom is wrong about A)/NQRWPE. It is not, as LSJ would have it, "frequently [used] in a contemptuous sense". This is true only when the speaker knows the addressee and chooses to address him as if he were a stranger anyway (in this case I use "him" and "he" advisedly. An anonymous woman will be addressed with GU/NAI). This phenomenon, so D, runs counter to a general law posited in address studies that addresses to strangers when used to known interlocutors will be polite. D. never really suggests an explanation for this, but comes close when she quotes Golden [1985: 101] saying 'it was the division between active male citizens and everyone else which really mattered in Athens.' We might say that whereas in modern Western cultures strangers are conventionally treated with respect, the Athenians regarded every outsider as by definition lower in status than an Athenian citizen.5 So maybe that time machine is not such a good idea after all.
Chapter 4, Other Aspects of Address Usage, is awkwardly placed. It treats material that, if it hadn't been for the fact that D. already has fifty pages' worth of appendices, would have been appendix material. She gives a succinct overview of three issues that are really tangential to the book but, in part, have bothered scholars in the past or are simply interesting in their own right. The first of these issues is address frequency. What factors determine the presence or absence of addresses in speeches? Readers should be aware that absence of an address at the beginning of a speech can indicate anger; recurrence of address within speeches can mark a climax. D. (p.194) gives Thucydides' Melian dialogue as an example, where none of the speeches contains an address except the Melians' last statement. With exemplary honesty for a scholar who apparently sees herself confronted with a non liquet she adds, in between this example and another strong one, that "nevertheless, in Greek as in English, addresses can also occur for no apparent reason."
Section 4.2 deals with the position of addresses within the sentence, and makes the observation that the address as a way to get somebody's attention will be placed at the opening of a sentence, in contrast to addresses in the midst of conversation which will normally be, as Fraenkel demonstrated, postpositive. I note that Alice's addresses to the mouse are unobjectionable in this respect as well. Finally (sec. 4.3) there is the use of W)=, "a topic which has received more attention than it deserves." In fact it is virtually the only subject to do with addresses that has a sizeable bibliography, and much of this bibliography D. dismisses politely before she goes on to the "facts of the question" and concludes in another two pages that "the search for a meaning for W)= is probably futile." So much for lots of well-intentioned, but sadly misdirected energy.
In chapter 5 D. returns to the central issue of her book, the significance of terms of address, and here she contrasts the various forms that were successively described but not differentiated in chapter 3. For instance, chapter 3 will tell us that a husband can address his wife by her name or with "wife". Chapter 5 asks what the difference is between the two. D. calls her description in 3 an investigation into the meaning of the forms of address, but personally I don't see that as a clear demarcation between the material in the two chapters. The fact that there are alternative expressions available to a speaker in a certain situation will always prompt the question of what the difference is between the two. I would gladly accept those differences as part of the meaning of the terms in question, others might say that terms that can both be used in a certain situation therefore necessarily have the same 'meaning.' I would have preferred a more continuous discussion, with the material now presented in 5 assimilated in chapter 3.
This review has, as is the way with reviews, concentrated on giving a rough outline and pointing out where there is room for improvement in the eyes of this reader. In conclusion I should therefore make clear that it was a pleasure to browse and read D.'s book. It is impossible to do justice to the wealth of information contained in it and to the thoroughness of D.'s investigation in this review. I have looked in vain for errors and misprints. My only remaining wish is for more indices (of languages, of terms) in addition to the index locorum and index of addresses. The highly specific Table of Contents does fill this lacuna in part, but as it is, there is no easy way to find D.'s references to Latin, Modern Greek or Kannada, or to 'nominative-for-vocative' (pp. 5, 18, 23). Sadly, this is one of those books that probably will not see an enlarged paperback reprint -- so reader, get it for your library before it is too late!
1. I wish to thank my home institutions in the first half of 1997, the Center for Hellenic Studies in Washington, DC, and the University of Amsterdam, where this review was actually written.
2. Lewis Carroll, Alice's Adventures in Wonderland, chapter 2, 'The Pool of Tears'.
3. 'Form of address' is not simply jargon for 'vocative' because it is the function, and not the form of expression, which is the criterion for inclusion. 'Vocative' would exclude nominative forms used as addresses.
4. This review is even more biased in this respect and in that sense does not accurately reflect D.'s discussion, which does include women and 'other minorities'.
5. Brown & Gilman themselves cite (1960:256) Grand (1930) for fifteenth-century Italian literature where Christians "say T [that is, use informal pronouns, HD] to Turks and Jews and receive V [formal pronouns]" so the universality of this law is questionable, at least. However, I am aware that I am here equating 'stranger' with outsider, which may not be quite the way this 'law' was originally intended.