1 Recent editors such as Rowe and Robinson have assigned the last statement to the elder rather than the younger Socrates. Monique Dixsaut gives an interesting argument for assigning this speech to young Socrates in “Une politique vraiment conforme à la nature,” in Reading the Statesman: Proceedings of the III Symposium Platonicum, ed. Rowe, Christopher J. (Sankt Augustin, Germany: Academia Verlag, 1955), pp. 254–56 and n. 14.Google Scholar

2 Annas, Julia, “Introduction,” in Plato: Statesman, ed. Annas, Julia and trans. Waterfield, Robin (Cambridge: Cambridge University Press, 1995), p. x.Google Scholar

3 Campbell, Lewis, “General Introduction,” in The Sophistes and Politicus of Plato (Oxford: Oxford University Press, 1867), p. xliv.Google Scholar

4 Statesman 263e6, 274e, 275d4, 276c3–4, 276e. References to all Platonic dialogues except the Laws are to the text of Burnet, John (Platonis Opera, 5 vols. [Oxford: Oxford University Press, 1900–1902])Google Scholar, except as noted, though I have checked the readings of the new Oxford text of the Statesman (in Duke, E. A., Hicken, W. F., Nicoll, W. S. M., Robinson, D. B., and Strachan, J. C. G., eds., Platonis Opera, vol. I [Oxford: Oxford University Press, 1995]Google Scholar) with the help of David Robinson, “The New Oxford Text of Plato's Statesman: Editor's Comments,” in Reading the Statesman. References to the Laws are to the Greek text of the Budé edition: Édouard Des Places, ed. and trans., Les Lois Liures I–VI, in Platon: Oeuvres Complétes, vol. XI, pts. 1–2 (Paris: Société D'Édition “Les Belles Lettres”, 1951)Google Scholar; and Diés, A., ed. and trans., Les Lois Livres VII–XII, in Platon: Oeuvres Complétes, vol. XII, pts. 1–2 (Paris: Société D'Édition “Les Belles Lettres,” 1956)Google Scholar. All translations are my own, unless otherwise stated.

5 Benardete, Seth, The Being of the Beautiful (Chicago: University of Chicago Press, 1984), 3:73Google Scholar. On the failure of the conversation in the Statesman to become a real dialogue between the Eleatic Stranger and Young Socrates see Rowe, Christopher, “The Politicus: Structure and Form,” in Form and Argument in Late Plato, ed. Gill, Christopher and McCabe, Mary Margaret (Oxford: Oxford University Press, 1996)Google Scholar. Yet, as Rowe does not discuss, the Eleatic Stranger only reluctantly engages in dialogue at all, and only on the guarantee that his discussants will prove young and tractable (Sophist 217d). It would appear from this passage that the failure of the Eleatic Stranger to engage his young interlocutors in genuine dialogue is an intentional feature of Plato's art in the Sophist and the Statesman.

6 see Guthrie, W. K. C., A History of Greek Philosophy, vol. V, The Later Plato and the Academy (Cambridge: Cambridge University Press, 1962–1981), p. 168Google Scholar; and Lane, Melissa, Method and Politics in Plato's “Statesman” (Cambridge: Cambridge University Press, 1998), p. 20.CrossRefGoogle Scholar

7 The new wave of work on the Statesman includes the collection of papers edited by Rowe, Christopher, Reading the Statesman, henceforth RSGoogle Scholar; Rowe's text, with translation and commentary, Plato: Statesman (Warminster, England: Aris and Phillips, 1995)Google Scholar; Plato: Statesman, ed. Annas, and trans. Waterfield, Google Scholar; Rosen, Stanley, Plato's “Statesman ”: The Web of Politics (New Haven: Yale University Press, 1995)Google Scholar; and Cropsey, Joseph, Plato's World: Man's Place in the Cosmos (Chicago: University of Chicago Press, 1995)Google Scholar. Much of this literature is surveyed in McCabe, Mary Margaret, “Chaos and Control: Reading Plato's Politicus,” Phronesis 42 (1997): 94–117CrossRefGoogle Scholar. Melissa Lane's recent Method and Politics presents novel understandings of temporality and method in the Statesman; I have discussed her key results elsewhere (Bryn Mawr Classical Review 9 [1998]).

8 Klosko, , The Development of Plato's Political Theory (New York and London: Methuen, 1986), p. 197Google Scholar. Versions of this linear approach are taken by Saunders, Trevor, “Plato's Later Political Thought,” in The Cambridge Companion to Plato, ed. Kraut, Richard (Cambridge: Cambridge University Press, 1992)Google Scholar; Annas “Introduction”; Zeitlin, Irving M., Plato's Vision: The Classical Origins of Social and Political Thought (Englewood Cliffs, NJ: Prentice—Hall, 1993), p. 145Google Scholar; and Samaras, Thanassis, “When did Plato Abandon the Hope that Ideal Rulers Might Appear among Human Beings?” History of Political Thought 17 (1996): 109–113Google Scholar. Rowe, , Plato: Statesman, pp. 17–18Google Scholar, and Lane, Method and Politics, raise appropriate doubts.

9 Some general points in comparing the Statesman with the Republic and the Laws are made well in Gill, Christopher, “Rethinking Constitutionalism in Statesman 291–303,” in RS, pp. 301–304.Google Scholar

10 For Scodel, Harvey, Diaeresis and Myth in Plato's “Statesman,” Hypomnemata 85 (Göttingen, Germany: Vandenhoeck and Ruprecht, 1987)CrossRefGoogle Scholar, Plato intends the Eleatic Stranger's differences with Socrates as departures not merely from Socrates, but from the truth. Scodel contrasts the Eleatic Stranger with “Plato/Socrates.” Yet Scodel's “Plato/Socrates” must either be a reading of Socrates' statements as Plato's views, or must be a sophisticated interpretive construct built on Plato's Socratic works, a construct that Scodel does not develop. Paul Stern presents a more complex account wherein the Eleatic Stranger ascends to Socratism through the dialogue in “The Rule of Wisdom and the Rule of Law in Plato's Statesman,”; American Political Science Review 91 (1997): 264–76CrossRefGoogle Scholar. Yet as we shall see, the Eleatic Stranger's views, even at the end of the Statesman, differ radically from those of Plato's Socrates and Athenian Stranger.

11 The Eleatic Stranger in the Statesman occasionally echoes Socrates in the Theaetetus, though he was not present to hear Socrates speak. Compare Theaetetus 174d with Statesman 262d ff; and, as Campbell notes (“General Introduction,” p. xxiii) compare also Sophist 253c with Theaetetus 172d.

12 Cropsey, Plato's World. I will discuss the other dialogues of the “world” only insofar as they illuminate the intentions of the Eleatic Stranger and of Plato himself in the Statesman. Howland, Jacob, The Paradox of Political Philosophy: Socrates' Philosophic Trial (Lanham, MD: Rowman and Littlefield, 1998)Google Scholar, is a recent treatment of these two tetralogies that is helpful on many points.

13 Rosen, , Plato's “Statesman,” pp. 170–71Google Scholar, and passim. The Eleatic Stranger understands politics as an art of grasping the proper moment or kairos (Statesman 305d3–4; see Melissa Lane, “A New Angle on Utopia: The Political Theory of the Statesman,” in RS; and now also Lane, Method and Politics). In this understanding the Eleatic Stranger appears to agree with the view Socrates expresses in the Republic that no artisan can practice more than one art because each artful activity has its proper moment, and one who is engaged in one art will necessarily miss the proper moment of another (Republic 370ab). The philosopher-rulers of the Republic must themselves artfully grasp the proper moment for the begetting of children of the highest class, applying their knowledge of the nuptial number (546a-d). If the notion of the technique of politics is an Eleatic error, it is equally a Socratic error. If the passage about the nuptial number suggests that Socrates espouses a technical account of true politics only ironically, we ought to be open (pace Rosen) to a similar irony by the Eleatic Stranger.

14 Rosen claims to distinguish between the views he attributes to the Statesman and those of Aristotle (Plato's “Statesman,” p. ix). Yet that distinction is overshadowed by his invocation of the peculiarly Aristotelian understanding of phron→sis in explicating both the Eleatic Stranger's speeches and the lessons that, on Rosen's view, Plato intends us to learn from them.

15 Bodeüs, , The Political Dimensions of Aristotle's Ethics, trans. Garrett, Jan Edward (Albany, NY: State University of New York Press, 1993), p. 36.Google Scholar

16 Two suggestive treatments of the relation between Aristotle's inquiry into political life and political life itself are Bodeüs, Ibid.; and Davis, Michael, The Politics of Philosophy: A Commentary on Aristotle's “Politics” (Lanham, MD: Rowman and Littlefield, 1996).Google Scholar Bodeüs argues that Aristotle's inquiry into politics is intended to aid the legislator rather than the statesman or citizen simply (pp. 54–68, 75). He does not explain what precisely the legislator can learn from Aristotle's political science apart from the value of the theoretical life (p. 125). Davis is emphatic that “the city needs political philosophy,” but he seems to understand to be impossible or impolitic to expound that need openly or clearly (see pp. 31,56). These questions about Aristotle require more exploration than I can give them here. I mention them only to point out that Rosen explains the obscurities of the Statesman by reference to a teaching about the relation of political practice to political philosophy that is, despite appearances, quite obscure in itself.

17 Rosen's exegesis of the Statesman in some respects continues and in some respects revises the projects of his Nihilism: A Philosophical Essay (New Haven: Yale University Press, 1969)Google Scholar and The Limits of Analysis (New York: Basic Books, 1980)Google Scholar. The most important revision would appear to be Rosen's current claim that in the Statesman Plato explores the “problem of constructivism,” so that the “Quarrel between the Ancients and the Moderns” cannot be understood merely as a division between those who claim that we know what we see and those who claim that we know what we make (Plato's “Statesman,” pp. 15–16).

18 In posing the question of the Stranger's Eleatism, I am engaged in an investigation anticipated by Mara, Gerald, “Constitutions, Virtue, and Philosophy in Plato's Statesman and Republic,” Polity 13 (1981): 355–82CrossRefGoogle Scholar. This article differs from Mara's in that, first, I compare the Statesman not only to the Republic but also to the Laws. Second, my comparison is founded in novel interpretations of the Eleatic accounts of the rule of law and the role of speech in political life. Third, I explicate formal features of the dialogue, the paradigm of weaving through which the Eleatic Stranger explicates his political teaching, the dialogue's dramatic date and context, and its frequent apparent changes of subject and tone. Christopher Rowe's work brings fresh and interesting thought to bear on many details of the Statesman, but while he poses the question of the Stranger's Eleatism in the introduction to his translation and commentary (Plato: Statesman, p. 10), he makes no sustained effort to explore it there or elsewhere.

19 The first formalization of the theory of the second-best is Lipsey, R. G. and Lancaster, Kelvin, “The General Theory of Second Best,” Review of Economic Studies 24 (1957): 11–32.CrossRefGoogle Scholar

20 See 295a2; Griswold, Charles, “Politik→Epist→m→in Plato's Statesman,” in Essays in Ancient Greek Philosophy, ed. Anton, J. and Preus, A. (Albany, NY: State University of New York Press, 1989), p. 157.Google Scholar

21 Statesman 295ab; Morrow, Glen R., Plato's Cretan City: A Historical Interpretation of the Laws (Princeton: Princeton University Press, 1960; reprinted with a new foreword by Charles H. Kahn, 1993), p. 586Google Scholar; Strauss, Leo, “Plato,” in History of Political Philosophy, ed. Strauss, Leo and Cropsey, Joseph, 3rd ed. (Chicago: University of Chicago Press, 1987), p. 75CrossRefGoogle Scholar; cf. Dombrowski, DanielPlato's Philosophy of History (Lanham, MD: University Press of America, 1981), pp. 139–40Google Scholar. Paul Stern argues that laws must be general because they aim at the general good, not the particular good of any individual covered by them (“Rule of Wisdom and the Rule of Law,” p. 269). Yet ideally, even to achieve a collective goal the ruler with full knowledge would adjust his orders to the task and situation of each individual subject.

22 See Statesman 294a, 300c-e. As Lane writes in “A New Angle on Utopia,” p. 287: “There is nothing antithetical in practice between law and art, only in the claim to ultimate authority.” For an extensive treatment of this point see Cohen, Jonathan, “Rex Aut Lex,” Aperion 29 (1996): 145–61Google Scholar; and also Gill, Christopher, “Plato and Politics: The Critias and the Politicus,” Phronesis 2 (1979): 148–67CrossRefGoogle Scholar, p. 150 and n. 7. This point has, I think, been more controversial than the text warrants. A good city completely without laws in any sense could not contain a plurality of individual, separate, human beings, since it would presume that the statesman was never separated from any other person among the ruled. While critics since Aristotle have accused Plato of denying that the citizens must be a many (see Politics 1261a), we should not privilege even the most authoritative critiques over the evidence of Plato's text itself. Rosen distinguishes between the rule of phron→sis which is beyond laws and the “epistemic city” which is ruled by laws crafted according to the supposed techn→ of the statesman (Plato's “Statesman,” p. 167). Yet he provides no evidence that cannot be more simply explained by the distinction between the (second-best) rule of sovereign laws and the (ideal) rule of sovereign wisdom through laws, a distinction noted by numerous scholars including Morrow, Plato's Cretan City, pp. 586–89; and Strauss, “Plato,” p. 75.

23 The principal difficulty with the interpretation adopted here is that the Eleatic Stranger does not explain how inferior cities learn the laws of the best city, as Rowe notes (Plato: Statesman, pp. 230–31, in the commentary to 300c4–6). Rowe instead contends that “imitating the best constitution well means simply sticking to established laws, which is just what the best city itself will do under the situation which always obtains in the inferior ones—namely when there is no knowledgeable person present to show what changes should be made” (Ibid., p. 17). Yet if we adopt Rowe's interpretation, the whole discussion of the written orders of absent doctors and trainers would lack a political analogue. Formally or literarily speaking this is a more serious problem than the one he raises.

24 I cannot agree with Rowe that the laws that the Athenian Stranger describes are “the laws which the expert ruler of [the Statesman] would write down if he had to” (“Introduction,” in RS, pp. 27–28, n. 98). The expert rule of the Statesman has one prescription for all cities, while the legislator of the Laws has a best and a second best regime that differ in numerous fundamentals.

25 Note that whereas the Laws present an ethical argument based on our inability to resist doing wrong when not liable to account for our behavior, the Statesman presents a cognitive argument for the rule of law, based on our ignorance of the true art of politics. See Vlastos, Gregory, “Socratic Knowledge and Platonic Pessimism,” Philosophical Review 66 (1957):226–38CrossRefGoogle Scholar, p. 235, n. 25, collected in Platonic Studies, 2nd ed. (Princeton: Princeton University Press, 1981)Google Scholar; and also my “The Unity of the Virtues and the Limitations of Magnesia,” History of Political Thought 19 (1998): 125–41.Google Scholar

26 Kahn, Charles, “The Place of the Statesman in Plato's Later Work,” in RS, claims that Plato presents the same “ideal theory” in the Statesman, Republic, and Laws (p. 54)Google Scholar. As we have seen, however, the Eleatic Stranger in the Statesman and the Athenian Stranger in the Laws have completely different theories of the application of ideal theory to non-ideal situations. The Eleatic Stranger thus contradicts Timothy Shiell's variability thesis, that the best regime in a given situation varies, depending on the circumstances (see his “The Unity of Plato's Political Thought,” History of Political Thought 12 [1991]: 377–90Google Scholar). Shiell's method for reconciling the very different regimes of the Republic and the Laws into a unified Platonic political theory thus breaks down completely if applied to the Statesman.

27 see Statesman 291e ff.; Crosson, Frederick J., “Plato's Statesman: Unity and Pluralism,” New Scholasticism 37 (1963): 28–43, 30.CrossRefGoogle Scholar

28 See Benardete, , Being of the Beautiful, 3:84Google Scholar; also Griswold, , “Politik→Epist→m→” p. 153Google Scholar; Mishima, Teruo, “Courage and Moderation in the Statesman,” in RS, p. 310, n. 11.Google Scholar

29 Rowe is thus incorrect to claim that the laws of the best city could not be suitable for adoption by inferior cities because the best city and inferior cities have different aims (Plato: Statesman, Commentary to 297d5–8).

30 Translated from the new Oxford text after Rowe, Plato: Statesman.

31 Pace Barker, Ernest, Greek Political Theory: Plato and his Predecessors, 6th ed. (London: Methuen, 1960), p. 333.Google Scholar

32 See Statesman 271e8–272al; Narcy, Michel, “La critique de Socrate par l'Etranger dans le Politique,” in RS, p. 231.Google Scholar

33 Campbell, , “General Introduction,” p. iiGoogle Scholar.

34 Theaetetus 172d9–e4, 144c5–9; Cropsey, , Plato–s World, p. 32.Google Scholar

35 On the wealth of historical and legal detail in the Laws see Morrow, Plato's Cretan City.

36 See, e.g., Laws 626d, 642b–64a, 693d, 753a, 886c.

37 Sophist 216a, 237ab, 241d–242d, 258c–e.

38 Stanley Rosen writes that “the demystification of politics renders nugatory the difference between humans and other animals” (Plato's “Statesman,” p. 87). Yet the Eleatic Stranger's demystification of politics denies only the distinction between human and animal politics. On the Eleatic view, it is in virtue of their ability to abstain from politics to converse about higher things that humans are more than simply bestial.

39 Gills, , “Rethinking Constitutionalism,” in RS, p. 304.Google Scholar

40 Campbell, , “General Introduction,” p. iiGoogle Scholar.

41 I translate here following the new Oxford text.

42 In Aristotle's Historia Animalium, the crane, by virtue of its intelligence, is a political animal ruled by a monarch (488a12, 614b19).

43 Charles Griswold thus blames Young Socrates unfairly for the outcome of this diaresis (“Politik→Epist→m→,” p.144). It is the Eleatic Stranger, not young Socrates, who insists on submerging the rule over humans in the variegated forms of animal herding.

44 Cf. Benardete, , Being of the Beautiful, 3:96.Google Scholar

45 In the myth of the age of Cronus told in the Laws, human beings are ruled in cities by daimonic kings and magistrates (713c–714a), While here in the Statesman they live without cities or regimes, governed by the daimons in herds(271e). For the Athenian Stranger, evan divinely rational rule is still somehow political, while for the Eleatic Stranger the political is an insignificant aspect of actual, all-too-human governance.

46 Sophist 260a; Klein, Jacob, Plato's Trilogy: Theaetetus, the Sophist, and the Statesman (Chicago: University of Chicago Press, 1977), p. 64.Google Scholar

47 See Griswold, , “Politik→Epist→m→,” pp. 152,160.Google Scholar

48 Rosen, , Plato's “Statesman,” pp. 68,137.Google Scholar

49 SeeMara, , “Constitutions, Virtue, and Philosophy,” p. 363.Google Scholar

50 See pp. 70–71 above.

51 The statesman attempts to achieve this concord by implanting a true opinion (al→th→doxa) about “the beautiful and the just and the good things” in the citizens (309c5–6). Yet even this true and hopefully steadfast opinion is to be implanted only as a means to achieve concord; this opinion in souls is merely the divine bond that holds together the embodied web of the regime.

52 Compare Euthydemus 292c, Republic 456e6–7. Roslyn Weiss argues that “the weaver-paradigm is in essence Socratic” (“Statesman as ; Caretaker, Physician, and Weaver,” in RS, p. 222), but this ignores the distinction between the Socratic (or Athenian) concern for virtue as an end and the Eleatic concern with virtue as a means to a corporeal end.

53 Socrates uses the image of philosophy as “deep” in the Theaetetus. Parmenides, Socrates says, was a man of great depth (Theaetetus 184al), while those whose “mental wax” lacks depth (bathos) receive indistinct impressions and so are poor learners. In the Republic and the Laws, the science of depth, stereometry, serves as the final mathematical propaedeutic for philosophy. (Republic 528a–d; Laws 818a ff., 961c; cf. Statesman 299e).

54 259b9–10; Benardete, , Being of the Beautiful 3:77.Google Scholar

55 291d6–7, cf. 268c6, 274cl, 297e12–13.

56 Compare Mishima on “the lack of depth and precision” in the Eleatic Stranger's account of the virtues (“Courage and Moderation,” in RS, p. 311).

57 On the literary and mythical context of the Eleatic Stranger's paradigm of weaving as the art of compromise needed to unite factions and preserve the “social fabric” see Scheid, John and Svenbro, Jesper, The Craft of Zeus: Myths of Weaving and Fabric, trans. Volk, Carol (Cambridge, MA: Harvard University Press, 1996)Google Scholar. They explore the ubiquitous connection in Greek and Roman myth and ritual between weaving and marriage, a connection that the Stranger invokes in his account of the divine and human bonds of kingly weaving.

58 The point I make by singling out the disjunction between herd and individual nurture, present in both the Statesman and the Laws, Rosen makes more speculatively in commenting on the disjunction between wild and tame animals at 261d3 and following: “I regard it as not impossible that Plato wishes us to think about human beings who live in herds but are nonetheless wild in the sense that they cannot be genuinely governed through commands of the statesman” (Plato's “Statesman,” p. 26).

59 Cf. 261d3–9 with Laws 666e.

60 Socrates' conversations with Alcibiades might seem to partake of both these failings.

61 Sophist 234d–235a; Klein, , Plato's Trilogy, p. 30.Google Scholar

62 Cf. Apology 29d–30b with Statesman 286a.

63 Statesman 258cd; cf. Klein, , Plato's Trilogy, p. 148.Google Scholar

64 On the Eleatic Stranger's concern with good of private, philosophic communion as opposed to public or political communion see Mara, “Constitutions, Virtue, and Philosophy”.

65 Republic 345e–347e, 519c–520d; Laws 753a, 969cd; Strauss, Leo, The Argument and the Action of Plato's Laws (Chicago: University of Chicago Press, 1975), pp. 83,186.Google Scholar

66 With Benardete and Rowe (and pace Clark, Stephen R. L., “Herds of Free Bipeds,” in RS, p. 239)Google Scholar, I understand the “willing tending” to be voluntarily performed.

67 See Rosen, Stanley, Plato's “Sophist”: The Drama of Original and Image (New Haven: Yale University Press, 1983), p. 111.Google Scholar

68 Many scholars have contended that the true statesman of the Statesman is, in fact, a philosopher-ruler. The only argument for this contention I have found is due to Klein, who claims that the true statesman must have knowledge of “the precise itself” (Plato's Trilogy, p. 177; cf. Statesman 284dl–2), and that this knowledge is tantamount to knowledge of the good. Yet the Eleatic Stranger says that “the precise itself” is needed for demonstrating the existence of the mean (284b-d), that is to say, for philosophic inquiry. He does not say that it is relevant to the practice of statesmanship, any more than it is relevant to the practice of weaving. Both weaving and statesmanship presume the existence of the mean, so to understand their existence philosophically, the existence of the mean must be demonstrated. Yet we have perfectly adequate cloaks without philosopher-weavers, so the successful practice of the arts does not depend on the inquiry into the mean. In Socrates' account in the Republic, by contrast, the justice of the philosopher-rulers' actions and commands depends on their knowledge of the idea of the good.

69 Statesman 283b; see Owen, G. E. L., “Plato on the Undepictable,” in Exegesis and Argument: Studies in Greek Philosophy Presented to Gregory Vlastos, ed. Lee, E. N., Mourelatos, A. P. D., Rorty, R. M. (Assen, Netherlands: Van Gorcum, 1973), p. 351Google Scholar; cited by Cole, Eve Browning, “Weaving and Practical Politics in Plato's Statesman,” Southern Journal of Philosophy 29 (1991): 195–208, p. 206, n. 3.CrossRefGoogle Scholar

70 Cf. Lane, , “A New Angle,” p. 290.Google Scholar

71 Statesman 283ab; Klein, , Plato's Trilogy, p. 165Google Scholar; Benardete, , Being of the Beautiful, III:73Google Scholar; cf. Skemp, J. B., Plato's Statesman (New Haven: Yale University Press, 1952), p. 78.Google Scholar

72 Compare the long and repetitive Buddhist sutra about detachment from human suffering whose purpose is to detach us from suffering such as it describes, as expounded by Kaufmann, Walter in Critique of Religion and Philosophy (Garden City, NY: Doubleday, 1961), pp. 396–405.Google Scholar

73 The passage from the Sophist referred to in the text is the “clear indication” that Rowe misses, to prove that the Eleatic Stranger is putting forward only a charade of discovery in “The Politicus: Structure and Form,” p. 176. My remarks in this final paragraph were motivated in part by the “Afterthought” of Rowe's paper; and by Rosen, , Plato's “Sophist” p. 185Google Scholar, and Plato's “Statesman,” p. 99.

74 An earlier version of this article was presented at a panel on “Citizens and Statesmen: Ancient and Modern” at the 1996 Annual Meeting of the American Political Science Association, where the discussant, Romand Coles, and members of the audience gave useful comments. Clifford Orwin helped and encouraged me in revising, and the anonymous referees contributed numerous valuable suggestions. Aline Linden read Dixsaut's and Narcy's articles with me. During the period of this research I received financial support from a National Science Foundation Graduate Fellowship, a Social Sciences and Humanities Research Council of Canada Postdoctoral Fellowship, and a Metcalf Fellowship from Victoria College of the University of Toronto.