George Kennedy is a brilliant scholar of Greek and Latin rhetoric, but he also wrote a slim book, Comparative Rhetoric, that makes a better go than most studies at being a genuinely comparative analysis of rhetoric across pre-modern cultures. Beginning with a discussion of animal behavior and encompassing Aboriginal Australian, Native American, and Chinese sources among others, Kennedy attempts to identify areas of commonality and variation across the cultures. Much (all?) of the book is clearly debatable, even if I’m not in a position to contest most of his claims, but the debates are productive ones.

The obvious problem is that as of today, the imbalance of evidence and analysis in favor of Greek and Latin makes a disinterested assessment very difficult if not impossible. but I think Kennedy makes a good-faith effort to adopt as neutral a stance as he can, and evaluate “rhetoric” as the purposes of linguistic expression in society, rather than as any particularly Western discipline. It is vastly less Western in its focus than many studies that purport to deprivilege western culture while only reifying its centrality. Even when I think he is showing his biases, I appreciate at least the effort to be fair. As a preliminary study, which is what Kennedy declares it to be, I rate it highly.

The book is compact and worth reading in whole. I have grouped some key passages under several headings.

Definition of Rhetoric

Rhetoric is not, I think, just a convenient concept existing only in the mind of speakers, audiences, writers, critics, and teachers. It has an essence or reality that has not been appreciated. I shall argue in this book that rhetoric, in essence, is a form of mental and emotional energy.

Rhetoric, in the most general sense, may thus be identified with the energy inherent in an utterance (or an artistic representation): the mental or emotional energy that impels the speaker to expression, the energy level coded in the message, and the energy received by the recipient who then uses mental energy in decoding and perhaps acting on the message. Rhetorical labor takes place.

Persuasion can be achieved in several ways, but nature has favored the use of signs because it is less expensive of energy than is the use of force. Rhetoric may be regarded as a form of mental energy, sparked by an emotional reaction to a situation in which an individual feels threatened or perceives the opportunity to gain some advantage.

Epideictic [Non-Judicial, Non-Deliberative] Rhetoric in Birds

Ritualization accompanied by epideictic utterance is a feature of animal rhetoric as it is of human life. Although the display and singing of birds in the mating season can be regarded as deliberative rhetoric, bird song outside the mating season is often epideictic in that it reaffirms existing relationships with other birds.

Once facility begins to be acquired, a fledgling bird begins experiments by amplifying topics. Amplification involves repetition, variation, combination, and substitution of themes. Erasmus’s famous treatise On Copia is the fullest description of how this was traditionally done in the West, using tropes and figures. Bird song exemplifies many figures of speech that are based on sound patterns: anaphora, homoeoteleuton, paronomasia, and the like. It does not employ figures of thought such as rhetorical question, apostrophe, or irony. It uses tropes only in the most literal sense that acoustic patterns or syllables seem sometimes to be substituted for others. Since these units do not appear to have cognitive value, there is no metaphor or metonymy, but the ability to make substitutions is fundamental to any development of troping. I noted earlier that synecdoche (genus for species or species for genus) is present in the communication of vervet monkeys and that some other animals understand metonymy. The ability of birds to combine their themes into different songs is significant because it illustrates in nature the potential to combine sounds into words that is the basis of human speech.

Rhetoric’s Conservative Function in Traditional Cultures

Physically, a tjurunga is a flat stone with some painted markings on it. These were the most sacred objects in Australian religion and almost the only articles of private property. A tjurunga is identical with the totem but all tjurungas are more or less alike. Their neutral appearance makes them capable of taking on any meaning. They are the ultimate synecdoche.

The traditional cultures of the world are exceedingly conservative; they resist change. Fear of change is an important source of rhetorical energy. Just as the basic impulse for rhetoric in the individual derives from the instinct for self-preservation, so the most common function of rhetoric in traditional societies is preservation of their accustomed beliefs and way of life. Among the ways this is done is by attributing the institution of customs to divine or semi-divine authority figures, by stressing the antiquity, continuity, and consistency of the customs, and by seeking to authenticate them by the use of archaic language. Social control is secured by those in the contemporary society who have been initiated into the language and the mysteries of the tradition and can both present it in its allegedly authentic form. and interpret it to the public. Opportunity for manipulation for personal gain exists, but probably only becomes a factor when special situations or crises arise. Deliberation of a sort always occurs on the problems of the group, but even in a traditional society with a more developed polity than that of the Australians, decisions are usually taken by consensus. As Malinowski (1922:62) wrote of the Trobrianders, “there is hardly ever much room for doubt or deliberation, as natives communally, as well as individually, never act except on traditional and conventional lines.”

At the linguistic level, archaic aboriginal rhetoric seems to illustrate a stage of what might be called “proto-metaphor. ” By juxtaposing words, a series of overlapping images, perceptions, or emotions can be projected without differentiating their fields of reference. The absence of explicit simile is evidence of the lack of such differentiation. Most modern theories of metaphor seem to me too intellectual and cognitive to describe this process. Its source lies largely if not entirely in animistic personification: the literal belief in the identification of species of animals with human beings and the animate reality of natural objects, including mountains, lakes, and rocks

Catachresis, synecdoche, and metonymy played leading roles in the early development of language. That is to say, when need occurred for a specific term for something, the initial impulse was not to invent a new term, but to find something in the existing vocabulary to “abuse”-the literal meaning of catachresis–in its place. The chief options then became either the use of genus for species or species for genus (synecdoche), if either existed in the language, or the use of some term that is physically contiguous and might be taken as characteristic of the idea needing expression (metonymy).

The Primacy of Rhetorical Form and Context Over Logic

Logical argument is often only a minor factor in the persuasive effect of a traditional formal speech. The evidence the speaker provides to support his position is its consistency with the traditional wisdom of the society, transmitted through proverbs and mythical or historical examples from the past. Although a Merina or Balinese speaker seemingly undermines his personal authority by his ethos of humility, he also implies authority by his ability to use the appropriate code. Such a speech is difficult to answer in any polite and acceptable terms: It has social power. Often, a speech is followed by silence, which effectively means agreement. In its most extreme form, by avoiding use of rational argument formal speech can forestall any potential logical objections, allegations of inconsistency on the part of the speaker, or attempts at rebuttal ( Bloch 1975:21). There is nothing to rebut. Sometimes, however, another elder will speak in the same style with a somewhat different proposal, but aiming at compromise and consensus, which is likely to be accepted. On rare occasions someone will not allow the silence of consent to follow a speech and will even ridicule it. Bloch notes ( 1975:10) that the Merina take great pains to avoid being addressed in formal language, which is used on other occasions as well as in councils, if they are not willing to accept the results.

Formal Speech as a Hegemonic Mechanism

Judith Irvine ( 1979) has identified four aspects of “formality” that apply cross-culturally.

First, in contrast to informal speech–story telling, conversation, gossiping, and the like–formal speech increases code structuring by imposing special rules of style and delivery on the speaker.

Second, the conventions of the appropriate code are consistently maintained.

Third, formal speech involves a positional rather than a personal identity. The speaker speaks in a certain political or social role, performing a public function rather than advancing personal interest.

Fourth, a central situational focus emerges in formal speech: That is, it deals with important activities and the central actors within them, leaving aside trivial matters.

Traditional deliberative speech is usually polite, considerate of the feelings of others, and relatively unemotional. When attacks are made, veiled or indirect language is often used, which allows the victim to save face and protects the speaker from immediate reprisal. The authority of a speaker is increased in some contexts by an ability to use a special, formal language, which carries the collective values of the community and makes use of grammatical parallelism, alliteration, assonance, indirect allusions, and metaphor. The most common metaphors are personifications of forces of nature, animals, or physical objects. Members of a traditional society find a speech in formal language difficult to answer, since any response seems to reject communal values. Another speaker may, however, express a somewhat different point of view, usually in a respectful way, and seek compromise. When this process breaks down, as sometimes happens, it is apt to break down completely, with a quick resort to ridicule, insult, and even violence. Someone who is dissatisfied with an imposed consensus may resort to sorcery to counter its effects.

Formal oratory was a conservative force, preserving the moral and political values of the past and reinforcing class distinctions.

“Formal language,” required in ceremonial or official contexts, often has poetic features and often seems archaic. Archaism certifies the authenticity of the message by suggesting its conformity with beliefs of the past. Use of formal language has to be learned and is not available to everyone; it thus exercises social power of a conservative sort.

The human inclination to develop formal languages is one of many indications of the basically conservative function of rhetoric in human history. Formal languages are often archaic or revivals of what is regarded as the pure form of the language used in the past. They thus contribute to the preservation of other past values. The requirement to use them for serious discourse helps ensure preservation of the status quo on the behalf of those in power and limits the ability of marginal groups, untutored in elitist language, from effecting change.

Aztec and Greek Rhetoric

As León-Patilla (1985:40) observes, “In many of the huehuetlatolli we find attempts to inculcate in the people that it is the destiny of the nobles to keep and transmit the ancient wisdom, to carry the people on their shoulders, and to feed the gods with the blood of captives seized in the sacred war. Such ideas confirm not only that these discourses were the speeches of nobles but also that among their aims was the reinforcement of the status of the ruling group.” The tone of Aztec epideictic differs greatly from that of Greek, reflecting differences between the two cultures: Aztec speeches are often very critical of the addressee, whereas Greek epideictic is more often given to flattery; Aztec oratory is harsh, austere, and fatalistic, whereas Greek is frequently playful and humane.

Mesopotamian Written Rhetoric

The Mesopotamian cultures seem consistently to privilege writing over speech; doubtless oral traditions lie behind their myths and legends, but writing incorporated and erased them. While other cultures, even after attaining literacy, have honored oral skills, eloquence, and the figure of the orator, Mesopotamian literature, so far as I have found, never did so. The scribes triumphed over the speakers. Even the gods wrote letters ( Grayson 1984), something unknown in other cultures.

Failure of Traditional Rhetoric in the Egyptain “The Complaints of Khakheperre Sonb”

[O that I had] unknown phrases, sayings that are strange,
novel, untried words, free of repetition;
not transmitted sayings, spoken by the ancestors!
I wring out my body of what it holds in releasing all my words;
for what was said is repetition, when what was said is said.
Ancestor’s words are nothing to boast of,
they are found by those who come after.
Not one speaks who spoke, there speaks one who will speak,
may another find what he will speak!
Not a teller of tales after they happen, this has been done before;
nor a teller of what might be said, this is vain endeavor, it is lies,
and none will recall his name to others.
I say this in accord with what I have seen:
from the first generation to those who come after, they imitate what is past.
Would that I knew what others ignore, such as has not been repeated,
to say it and have my heart answer me, to inform it of my distress,
shift to it the load on my back, the matters that afflict me,
relate to it of what I suffer, and sigh “Ah” with relief!
(ADAPTED FROM LICTHEIM 1973, I:146-47)

Hebrew Biblical Rhetoric

Lamentations and prophecies are found in Mesopotamian, Egyptian, and other cultures, as noted earlier, but the texts of the Hebrew prophets exhibit a far more complex rhetorical invention than is found elsewhere. This results primarily from two factors, the nature of Jewish religion and the editing and publication of written texts. The religion of the Jews as it developed into monotheism, with all power assigned to a god who claimed to bestow special favor on his chosen people, created an unprecedented theological and rhetorical problem when that favor seemed to be withheld.

Chinese Rhetoric

Like traditional rhetoric in other parts of the world, Chinese rhetoric as we see it in the earliest texts was conservative, even reactionary, aimed at consensus, and sought to reaffirm social and political hierarchies, modeled on family relationships in which great emphasis was put on the authority of a father over his sons and the respect of a son for a father.

Chinese thinkers, however, often emphasized a need for frankness and sincerity in political contexts to a degree not found elsewhere, and during the long history of China rhetorical teaching was given considerably greater depth and detail than is found in the Near East. This is primarily because of the rich philosophical traditions, especially Confucianism, that came to dominate Chinese thought. Neither Confucius nor most other Chinese thinkers held a very high opinion of the intelligence of the general public; what they have to say about speech, persuasion, and other aspects of rhetoric is addressed to rulers or to their own philosophical students and does not consider techniques of addressing a mass audience. It is equally true that most ancient Chinese speeches are addressed to a single individual, a ruler or one of his ministers, or to a small group of political advisers ( Garrett 1993b:22-23).

Although metaphor is constantly present, imagery is not rationalized as simile: “Early Chinese songs do not as a rule introduce a comparison with an ‘as if’ or ‘like,’ but state it on the same footing as the facts they narrate” ( Waley 1960:13). This is consistent with the treatment of metaphor in traditional oral cultures as I have described it in earlier chapters.

Chinese rhetoric as seen in the Shu generally avoids pathos except in military exhortations and in some of the more severe announcements or instructions. It is strong in ethos–the authority and character of the speaker, the tradition of the ancestors who continue to watch the living, the moral rightness of the message—but it is also not lacking in logical argument; as the Viscount of Ke said in an instruction to King Chou, “the virtue of speech is accordance with reason” ( Legge III:326). Argument is chiefly inductive, based on examples, precedents, quotation of authorities, and analogies. Deductive argument in the form of enthymemes seems undeveloped. Speeches are often clearly divided into parts that perform the functions of proemium, narration, proof, and epilogue.

Some important concepts in Confucian thought are the Tao (the “Way”), the proper course of human conduct based on the model of antiquity; te, often translated “virtue,” a person’s potentiality to act in accordance with the Way; li, ceremony or good manners; yi, the conduct fitting one’s role or status; and jen, used as a noun to mean “noble” and as a verb to mean “be civilized” in conduct ( Graham 1989:11-19). The mind/body dichotomy of Western thought was, happily, unknown to the Chinese; jen is not an inner state. Fingarette ( 1972:55) claims it was conceived as “a directed force operating in actions in public space and time, and having a person as initial point-source and a person as the terminal point on which the force impinges.” It thus closely resembles “rhetoric” as a form of energy, a definition I suggested earlier in this book.

Chinese and Greek Sophistry

Sophistry in Greece arose in the fifth century BCE, a time of conflicting philosophical doctrines claiming knowledge of truth. The sophists were skeptical of these claims, which they undermined by verbal subtleties of argument and the use of paradox; they sought to teach practical ways of success in the world based primarily on techniques of public speaking. They were much interested in language; indeed, the sophists began the study of grammar and philology. Although Athens became the center of the sophist movement, most of the leading sophists were not Athenians but itinerant teacher-orators who supported themselves by traveling about the Greek world, giving demonstrations of their skills and offering political advice to Greek states. Something analogous to Greek sophistry had appeared in India a century or two earlier; in China it may have begun as early as the sixth century but flourished especially in the fourth and third centuries, thus overlapping the period of sophistry in classical Greece.

Sophistry in some form seems to be a regular development in “sophisticated” literate societies when political, social, and moral conditions are undergoing change, conflicting philosophical schools arise, each claiming access to truth in a different way, and individual teachers appear who are independent of state bureaucracies and offer advice to rulers. All these conditions existed in China in the fourth century. What Graham ( 1978:15-18) has called “the metaphysical crisis of the 4th century” seems especially to have contributed to a sophistic turn in Chinese thought. This was precipated by the teachings of Yang Chu, who argued for making judgments on the basis of the interests of the individual and introduced the concept of hsing into Chinese philosophy. Hsing is human nature, “the spontaneous tendency of the living organism throughout its life span … we obey Heaven, not as Confucians and Mohists suppose by behaving morally, but by nurturing and harmonising the vital tendencies and spontaneous inclinations which Heaven instilled in us when we were born” ( Graham 1978:16-17). This might be said to include rhetorical energy.

Abstraction in Indian Rhetoric

As the discussion here will reveal, Indians put a very high value on speech, higher perhaps than that found in any other ancient culture, and in marked contrast to Mesopotamians. Even long texts were memorized and transmitted orally, and knowledge of sacred texts was guarded by priests; some may have been written down in the sixth or fifth centuries, many probably did not exist in written form until the fourth or third century, and our texts may reflect conditions of that time or have been significantly affected by the thought processes involved in writing (Goody 1987:110-22).

Despite the resistance to writing, abstract thought developed in India at an early time to a greater extent than in any other culture I have so far discussed. Why this should be so is difficult to say. It may have been facilitated by the ability of Indo-European languages to coin abstractions, which is also evident in the development of Greek philosophy. An important factor was the existence of a social class–the Brahmin priests–with the leisure for contemplation. An inclination to abstraction probably reflects some basic feature of Indian religious feeling. Certainly it is evident first in religious texts of the classical period, which are filled with definitions and subdivisions of abstract concepts. It is also evident in early Indian scholarship on law, grammar, literature, and science.

Rhetorically, as well as geographically, India lay between China and Greece. In both India and China discussions of speech in the archaic and classical periods are found in a context of political and ethical thinking, not set off as a separate discipline, as happened in Greece. In all three countries logical argument was developed as a subtle tool, leading to a form of sophistry. There is more inclination to classification of abstract concepts and more explicit celebration of the power of speech in India, seen for example in the Hymn to Vak or Yajnavalkya’s tribute to Brahma as speech, than in China. Both Indian and Chinese thinkers conceptualized some aspects of rhetoric and created a terminology for criticism, the Indians in greater detail than the Chinese.

Simile vs. Metaphor

At the linguistic level, archaic aboriginal rhetoric seems to illustrate a stage of what might be called “proto-metaphor. ” By juxtaposing words, a series of overlapping images, perceptions, or emotions can be projected without differentiating their fields of reference. The absence of explicit simile is evidence of the lack of such differentiation. Most modern theories of metaphor seem to me too intellectual and cognitive to describe this process. Its source lies largely if not entirely in animistic personification: the literal belief in the identification of species of animals with human beings and the animate reality of natural objects, including mountains, lakes, and rocks.

Unlike synecdoche and metonymy, metaphor carries, or can carry, felt emotion. Simile would seem to be a rationalization of metaphor: that is, use of an explicit comparison when the speaker realizes a metaphor would not be literally true or would seem far-fetched. Proverbs function like metaphors; that is, they are substitutions or transferences of a traditional saying that takes the place of a more specific description of the immediate situation. Appeal to proverbs is often a feature of the code of politeness cultivated in traditional societies and thus a rhetorical strategy.

The hymns of the Rigveda contain references to the existence of councils and assemblies among the Aryans and a few use a simple dialogue form. Like other religious texts, they are frequently metaphorical in expression, but unlike other early or traditional poetry we have met they also often specify that something is “like” something else rather than seeming to identify two concepts. Since this is also a feature of early Greek poetry, it is tempting to think that there was something about the experience of Indo-European tribes that caused them to distinguish what was in some way “like” something else from what could be identified with something else in an undifferentiated way. Could it be that in their wanderings over vastly different landscapes, encountering different people with different ways of life and different beliefs, an experience in scale unlike that of other people I have discussed, they observed phenomena with which they could not entirely identify but which had some likeness to something they knew and found a linguistic way to express this intermediate stage of the same and the other? A contributing factor may have been the ambiguity of the Indo-European verb “to be” in all its forms. Only in Indo-European languages (and in ancient Sumerian where simile also appears) does a single verb function both as copula, which predicates some property or attribute (e.g., “This object is red”), and also means “to exist” ( Dewart 1989:259-300). Thus an English phrase such as “The god is a red-burning fire” can mean that the god exists and really is a red fire and that the god has some attribute of a red fire.

Atypical Aspects of Greek Rhetoric

By the fifth century, however, as seen in the works of Aeschylus, Pindar, and their successors, Greek poetry abounds in metaphor. Unlike other literary traditions, Greek thus seems to reverse the usual development: instead of movement from an early inclination for metaphorical expression to an increased literalism, we find a movement from clear, relatively literal expression toward increased exploitation of metaphorical imagery.

The most distinctive feature of Greek public address in contrast to that of many other cultures is its eristic qualities. In the traditional oral and early literate societies I have described earlier, the goal of deliberative rhetoric is usually consensus and concord in accordance with conservative values, and sharp altercation is avoided if possible. Differences are usually politely or indirectly stated. In Egypt, Palestine, India, and China there are injunctions to turn away wrath with a soft answer, or even to be silent; this was not the attitude of the Greeks.

Strife among human beings in Greece might be thought to receive some validation from strife among the gods. What is unusual in Greece is the acceptance, even the celebration, of contention and rivalry in civic society. With the canonization of the Homeric poems as the cultural textbooks of the Greeks, the immorality of the gods in epic became a problem for ethical philosophers.

The history of classical Greece is the history of the rivalries, plots, and wars of proud, independent city-states. Although Greeks ordinarily valued family ties and friendship, their contentiousness at times, in strong contrast to Chinese culture, even countenanced revolt of children against parents or other family members. Orestes killed his mother; in the story of the Seven against Thebes brother killed brother. Aristophanes’s comedy, Clouds, portrays a son out-arguing and even beating his father.

The Western world is indebted to the Greeks for the earliest models of democracy, but Greek democracies were almost always at crisis stage, riven by faction, and easily degenerated into mob rule and sometimes civil war under the influence of demagogues.

The acceptance of majority decision, even a majority of one, has significant effect on rhetorical practice. If a speaker does not need to secure consensus, he need not try to conciliate the more extreme opponents, can largely ignore some of their concerns, and can concentrate on solidifying support with those already inclined to agree and winning over the doubtful. Vigorous, even personal, attack on opponents and their motives contributes to this end.

The Greeks delighted in contentious argument; they often put a relatively low priority on telling the truth if a lie would be more effective; slanderous invective was not out of order in a court of law. Perhaps because of this, they seem to have become more tolerant of blatant flattery than most egalitarian cultures. Autocratic societies are, of course, another matter, and the existence of wealthy individuals can encourage flattery, as in the case of Hausa “roko” mentioned in chapter 4. When Socrates in Plato’s Gorgias (463a9) is pressed to offer his own definition of rhetoric he calls it a form of flattery. What he has in mind are primarily the efforts of democratic politicians (the rhetores) to gain influence by praising an audience and telling them what they want to hear rather than what is wise and just. Even the morally austere Aristotle suggests topoi for flattery in public address.

It would doubtless be an exaggeration to say that speakers in other cultures do not understand logical contradiction, but it is perhaps true that Western contentiousness tends to identify and sharpen contradictions. In other cultures, and now in poststructural thought in the West, there is a greater inclination to entertain the possibility that two seemingly contradictory statements may both be true in some sense; for example, if a term is used metaphorically in one of the statements. Yang and yin in Chinese thought are complementaries, not opposites; Mencius’s doctrine of multiple definitions is a Chinese example of a different form of reasoning. Western thinking, beginning with the Greeks, has tended to polarize truth and fiction, good and bad, body and soul, conservative and liberal, and other such concepts, for the sake of clarity but often unnecessarily.

The Contentious Bit

First in biological evolution, then in human cultural evolution the communicative instinct has been generalized to a desire to secure not only the necessities of life, but also what is perceived as the qualities of a good life. The basic function of rhetorical communication is defensive and conservative; but to secure or preserve the quality of life for one individual or one group may seem to require offensive actions and efforts at change. Traditional human societies have been strongly resistant to change, which is usually perceived as change for the worse. The major function of rhetoric throughout most of human history in most of the world has been to preserve things as they are or to try to recover an idealized happier past. In times of-stress, the latter sometimes takes the form of millenialism, prophecy of the return of a Golden Age or the coming of a messiah. Occasionally, influential individuals have undertaken rhetorical programs for change; examples cited in earlier chapters include Akhenaton in fourteenth-century BCE Egypt, Moses as described in the Old Testament, Solon in sixth-century Greece, Ashoka in India, and the Legalists in China. Popular belief in the possibility of progress and thus openness to change for the better is largely limited to classical Greek and modern Western societies and even there often resisted.

The function of deliberation in traditional societies is the achievement of consensus: not the acceptance of the view of a majority but explicit or tacit unanimity. The pressure for consensus is so great that if the process breaks down, as it sometimes does, open fighting may occur, dissenters may move away, or may be forcibly silenced. Lack of unanimity is a threat not only to leaders but to the maintenance of society. Non-Western societies that have accepted Western democracy often continue to try to impose uniformity of public opinion in a way disquieting to Westerners; Singapore is a good example. The appearance of uniformity remains an important value in China and many Third World countries. Comparative study of rhetoric helps understanding of why this inclination is natural.

The other ancient society in which open contention is common is that described in the ancient Indian epic, Mahabharata. Since the ancient Greeks and the Aryan invaders of India derived their language and some of their cultural institutions from the same ultimate source, the Indo-Europeans, this may be a heritage from contentious tribesmen who once roamed the steppes of Russia.

It is that final section that will raise eyebrows, and I do think that Kennedy has drawn too sharp a distinction between western and non-western societies. I believe he overemphasizes contention as a causal factor. I suspect he is right that the western rhetoric as derived from Greek and Roman sources is atypically contentious, with both positive and negative implications: more opportunities for disagreement, but more opportunities for pointless conflict. But contention and argument are amorphous things, and I am not at all certain that open disagreement necessarily entails a lack of consensus at the deepest level. The case of India, where Kennedy unconvincingly appeals to a common Indo-European origin to explain the amount of contention, also makes me think that he has overstepped the evidence there.

But I feel in greater agreement with Kennedy’s assessment of uniformities. The overall message is rather grim and Foucauldian: language has developed as a force primarily for enforcing hierarchical social roles and preserving an existing power structure, creating a vicious circle in which power reinforces the rhetoric which reinforces that same power. This is seen  most clearly in the establishment of formal speech, restricted to an elite and set out as a clear demarcation of status.

Rhetoric does, however, open up possibilities for rebellion, subversion, creativity, and genuine debate, as beautifully shown in “The Complaints of Khakheperre Sonb.” Rhetoric is a conservative force that, in its flexibility and creativity, allows for innovation and dissent, only to lash back at that innovation as a result of its conservatism. Literacy only amplifies the possibilities for instability.