Autonomy and hierarchy

Social Philosophy and Policy

Volumn 20, Issue 2, 2003, Pages 156-176

  Bratman, Michael E ^a  

^a STANFORD UNIVERSITY   (United States)

Author keywords

[No Author keywords available]

Indexed keywords

EID: 26444444344     PISSN: 02650525     EISSN: 14716437     Source Type: Journal    
DOI: 10.1017/S0265052503202065     Document Type: Review

References (81)

1. See Harry G. Frankfurt, The Importance of What We Care About (Cambridge: Cambridge University Press, 1988).
2. See also Gerald Dworkin, "Acting Freely," Nous 4 (1970): 367-83;
3. Wright Neely, "Freedom and Desire," Philosophical Review 83 (1974): 32-54;
4. and Keith Lehrer, "Freedom, Preference, and Autonomy," The Journal of Ethics 1, no. 1 (1997): 3-25.
5. Frankfurt, "Freedom of the Will and the Concept of a Person," in Frankfurt, The Importance of What We Care About, 11-25.
6. For an important, early response to Frankfurt's original essay, see Gary Watson, "Free Agency," Journal of Philosophy 72 (1975): 205-20. Watson offers an alternative approach, one that replaces appeal to motivational hierarchy with an appeal to a distinction between motivational and evaluative orderings. Watson also points to at least two potential concerns for the hierarchical approach: (1) a concern about the grounds for seeing higher-order desires as having a stronger claim to speak for the agent than do lower-order desires, without embarking on an unacceptable regress; and (2) a concern about the idea that, in deliberation, we reflect on our desires rather than directly on our options. I discuss this second concern below, in the main text of this essay.
7. Michael E. Bratman, "Identification, Decision, and Treating as a Reason," in Bratman, Faces of Intention (New York: Cambridge University Press, 1999), 185-206.
8. Here I am, broadly speaking, following both Frankfurt and J. David Velleman. See, in particular, J. David Velleman, "What Happens When Someone Acts?" in Velleman, The Possibility of Practical Reason (Oxford: Oxford University Press, 2000), 123-43. In speaking of functioning that realizes such an abstract property, however, I am making room for the possibility of multiple realizations. I am unsure whether Frankfurt or Velleman would also want to do so. (My appeal in the text to a "central kind of functioning" signals that my concern is with the limited claim that one theoretically important realization involves motivational hierarchy.) Let me also note here that, as I understand the notion of functioning, not all causal impacts will be included in an attitude's functioning.
9. In Michael E. Bratman, "Reflection, Planning, and Temporally Extended Agency," Philosophical Review 109, no. 1 (2000): 35-61, I explore the role, in strong forms of human agency, of higher-order policies concerning the functioning of first-order desires in one's motivationally effective practical reasoning. A number of individuals have asked whether such policies about practical reasoning need to be higher order. (Samuel Scheffler once raised this question in a particularly helpful way in correspondence.) The present essay responds to these concerns.
10. Watson, "Free Agency," 205-20.
11. Ibid., Watson, "Free Agency", 219.
12. This is a central Frankfurtian theme. The idea of casting this problem together with the problem, noted below in the text, of underdetermination by value judgment parallels aspects of Marth C. Nussbaum's discussion in her The Fragility of Goodness: Luck and Ethics in Greek Tragedy and Philosophy (Cambridge: Cambridge University Press, 1986), chap. 4.
13. In "A Desire of One's Own" (Journal of Philosophy, forthcoming), I note several different ways of interpreting this constraint of intersubjectivity. We might, for example, see a judgment of value as made from a Humean "common point of view," or as a judgment that those who are appropriately rational and informed would converge in a relevant way, or as involving the expression of a demand on others to converge in relevant ways. And other interpretations are possible. For our present purposes we do not need to settle on a specific interpretation, though for ease of exposition I will sometimes write in ways that fit most naturally with the second of these interpretations.
14. For a version of this second interpretation see Michael Smith, The Moral Problem (Oxford: Basil Blackwell, 1994), 151-77.
15. A number of philosophers have emphasized ways in which such judgments of value can underdetermine the specific contours of an individual life. For present purposes I will take it for granted, without further argument, that there frequently is some such underdetermination. See, e.g., Isaiah Berlin, Four Essays on Liberty (Oxford: Oxford University Press, 1969);
16. Robert Nozick, Philosophical Explanations (Cambridge, MA: Harvard University Press, 1981), esp. 446-50;
17. and Joseph Raz, The Morality of Freedom (Oxford: Oxford University Press, 1986), chap. 14. Consider also T. M. Scanlon's remark that "one cannot respond to every value or pursue every end that is worthwhile, and a central part of life for a rational creature lies in selecting those things that it will pursue."
18. T. M. Scanlon, What We Owe to Each Other (Cambridge, MA: Harvard University Press, 1998), 119.
19. Jean-Paul Sartre, "Existentialism Is a Humanism," in W. Kaufmann, ed., Existentialism from Dostoevsky to Sartre (1956; reprint, rev. and expanded, New York: Meridian/Penguin, 1975), 354-56.
20. My discussion throughout this essay assumes the approach to intention that I have called "the planning theory" and that I present in Michael E. Bratman, Intention, Plans, and Practical Reason (Cambridge, MA: Harvard University Press, 1987; reissued by CSLI Publications, 1999). I discuss policies, esp., at 87-91.
21. also discuss policies in Michael E. Bratman, "Intention and Personal Policies," Philosophical Perspectives 3 (1989): 443-69.
22. Concerning this qualification see my discussion of what I call "quasi-policies" in Michael E. Bratman, "Reflection, Planning, and Temporally Extended Agency," 57-60. In most of my discussion here I will not keep repeating this qualification (though I will return to it briefly below in note 51).
23. Cp. Nozick, Philosophical Explanations, 446-49.
24. The basic idea of creature construction comes from Paul Grice. Grice aimed to "construct (in imagination, of course) according to certain principles of construction, a type of creature, or rather a sequence of types of creature, to serve as a model (or models) for actual creatures." See Paul Grice, "Method in Philosophical Psychology (From the Banal to the Bizarre)," Presidential Address, Proceedings and Addresses of the American Philosophical Association 68 (1974-75): 37.
25. My discussion is in Michael E. Bratman, "Valuing and the Will," Philosophical Perspectives 14 (2000): 249-65.
26. For some intermediate steps in this construction see "Valuing and the Will," 252-57.
27. In the central case that I consider in "Valuing and the Will," the self-governing policy concerns first-order motivation that is already present. I also note, however, that there can be cases in which the policy involves, rather, a commitment to acquiring certain desires; and such a policy might concern one's treatment of certain desires, were one to acquire them. For a related but different conception of a connection between valuing and policies, see David Copp, Morality, Normativity, and Society (New York: Oxford University Press, 1995), 177-78.
28. For a closely related distinction see Philip Pettit and Michael Smith, "Backgrounding Desire," Philosophical Review 99 (1990): 565-92. In what follows, my first case corresponds to cases in which, in their terminology, the desire is in the "foreground." My second case is similar to one kind of case in which, in their terminology, the desire is in the "background."
29. We might also see (a) as alluding to further conditions that the desire fulfills, for example, that it is a considered desire.
30. Appeal to an evaluative expression of the desire is characteristic of Donald Davidson's views about practical reasoning. See, e.g., Donald Davidson, "Intending," reprinted in Donald Davidson, Essays on Actions and Events (Oxford: Oxford University Press, 1980), 85-86. John Cooper emphasizes how, on Aristotle's theory of virtues of character, even appetites and forms of anger and grief involve judgments about the good or what ought to be done, although these judgments are not themselves based on reasoning that aims at determining what is good or what ought to be done. Cooper also emphasizes the permanence of these nonrational desires even in a human being of Aristotelian virtue of character.
31. See John M. Cooper, "Some Remarks on Aristotle's Moral Psychology," reprinted in Cooper, Reason and Emotion: Essays on Ancient Moral Psychology and Ethical Theory (Princeton, NJ: Princeton University Press, 1999), 237-52.
32. We might try to see (b), when it is an expression of (a thought involved in) my desire, as sometimes involving an implicit indexical element: Revenge is a justifying consideration (from my point of view). We would then need to address the broadly Frankfurtian issue of which point of view is mine. This is the issue of agential authority that I turn to briefly below in Section V. A consequence of the approach to agential authority sketched in Section V (see also note 39) is that there are desires that are not appropriately expressed in this way.
33. Let me note two complexities. The first concerns Model 2. In some cases the desire for X will, even prior to an endorsing policy, already involve a thought of X as a justifying consideration, or will at least be plausibly expressible along the lines of (b). But there are, I think, also cases which do not fit well into such a picture: for some cases of pre-reflective anger, for example, this will seem to be an overly intellectualistic picture. Nevertheless, if in a case of this latter sort one does arrive at a self-governing policy in support of treating the anger as reason-providing, then this policy may infuse or shape the anger so that it becomes (or involves a thought that is) expressible in this way. So the reasoning supported by the policy can be Model 2 reasoning. A second complexity concerns motivation in the absence of either kind of practical reasoning. An agent who rejects her desire for revenge has a self-governing policy of not allowing that desire to lead to action by way of Model 1 or Model 2 practical reasoning. I think we can also suppose that the agent's policy rejects an effective motivational role for that desire, even if that role does not involve such practical reasoning-perhaps the desire of a Frankfurtian "unwilling addict" could in some cases motivate action in this latter way. However, it is policies specifically about the roles of desires in motivationally effective practical reasoning that are central to autonomous action; or so I will be claiming below in the text. These policies will be my main concern here.
34. Bratman, "A Desire of One's Own."
35. For such talk about the "shape" of our lives see Charles Taylor, "Leading a Life," in Ruth Chang, ed., Incommensurability, Incomparability, and Practical Reason (Cambridge, MA: Harvard University Press, 1997), 183.
36. I discuss this idea further in Michael E. Bratman, "Shared Valuing and Frameworks for Practical Reasoning," in R. Jay Wallace, Philip Pettit, Samuel Scheffler, and Michael Smith, eds., Reason and Value: Themes from the Moral Philosophy of Joseph Raz (Oxford University Press, forthcoming). Note that the idea is not that such policies directly change what is valuable - though there is room for an indirect impact by way of the value of living in accord with such policies, once they are adopted.
37. For this distinction see David Lewis, "Dispositional Theories of Value," in Lewis, Papers in Ethics and Social Philosophy (New York: Cambridge University Press, 2000), 68-94;
38. Gary Watson, "Free Action and Free Will," Mind 96 (1987): 150;
39. and Gilbert Harman, "Desired Desires," in Harman, Explaining Value and Other Essays in Moral Philosophy (Oxford: Oxford University Press, 2000), 117-36, esp. 129-30.
40. For such talk of self-creation see Raz, The Morality of Freedom, 385-90;
41. and Joseph Raz, "The Truth in Particularism," in Raz, Engaging Reason: On the Theory of Value and Action (Oxford: Oxford University Press, 1999), 242-45.
42. See Cooper, "Some Remarks on Aristotle's Moral Psychology," esp. 247-50, where Cooper highlights the contrast with the Stoics.
43. Allan Gibbard, Wise Choices, Apt Feelings (Cambridge, MA: Harvard University Press, 1990). While I think that Gibbard's focus on issues about social coordination is of great importance, I see my discussion here as neutral concerning the debate between Gibbard's expressivist understanding of value judgment and certain more cognitivist approaches. This is part of an overall strategy - a kind of method of avoidance, to use John Rawls's terminology - of trying to articulate important structures of human agency in ways that are available to a range of different views in metaethics.
44. I think that we can also expect forms of shared valuing (in contrast with shared judgments of value) to emerge. See Bratman, "Shared Valuing and Frameworks for Practical Reasoning."
45. I think that this is implicit, for example, in Gibbard's effort to distinguish between an "existential commitment" and accepting "a norm as a requirement of rationality." See Gibbard, Wise Choices, Apt Feelings, 166-70.
46. Which is not to say that these self-governing policies may not themselves be responsive to the creature's judgments of value.
47. A fuller discussion also would consider both "quasi-policies" (see note 14 above) and "singular commitments." See Michael E. Bratman, "Hierarchy, Circularity, and Double Reduction," in Sarah Buss and Lee Overton, eds., Contours of Agency: Essays on Themes from Harry Frankfurt (Cambridge, MA: MIT Press, 2002), 65-85. These complexities can be put to one side here, however, since our primary concern is with a kind of hierarchy involved in all of these phenomena.
48. I consider in the text below, in Section VI, the objection that there may be a gap here between function and content.
49. See Frankfurt's work on wholeheartedness in, for example, Harry G. Frankfurt, "The Faintest Passion," reprinted in Frankfurt, Necessity, Volition, and Love (Cambridge: Cambridge University Press, 1999), 95-107.
50. See note 5 above.
51. Concerning these two kinds of authority, see Michael E. Bratman, "Two Problems About Human Agency," Proceedings of the Aristotelian Society 101 (2001): 309-26.
52. I expand on these matters, and their relation to ideas about personal identity, in Bratman, "Reflection, Planning, and Temporally Extended Agency." In pages 48-51 of that essay I describe the cited nonconflict condition as a version of what Frankfurt calls "satisfaction." In my discussion of higher-order policies (below in the text) I will take it for granted that some such satisfaction condition is realized. A full account of satisfaction would also need to consider the significance of conflict with singular commitments concerning what to treat as justifying (see note 33 above).
53. We might say that such self-governing policies help constitute the agent's justificatory point of view. So if such a self-governing policy were to reject a desire for X, and that desire were nevertheless to involve the thought that X is a justifying consideration from that agent's point of view, that thought would be false.
54. Velleman, "The Guise of the Good," in The Possibility of Practical Reason, 99-122.
55. Cp. Velleman, "Introduction," in The Possibility of Practical Reason, 11.
56. On this account autonomous action is compatible with the persistence of first-order motivation that diverges from what is supported by one's self-governing policies. Within the proposed model, what autonomy requires is that one's self-governing policies actually do guide one's relevant reasoning and action. Further, there can be cases - e.g., our case of principled sexual abstinence - in which one's self-governing policy rejects a desire for X even though one acknowledges the value of X.
57. See Susan Wolf, Freedom Within Reason (Oxford: Oxford University Press, 1990);
58. but see also Gary Watson, "Two Faces of Responsibility," Philosophical Topics 24 (1996): 240. Relatedly, we might also consider a constraint that, at the least, the relevant self-governing policies not favor one's own loss of autonomy or complete domination by others. Here, again, we need not settle the issue in order to argue for the AH thesis.
59. As I understand her views, this is a theme in Christine M. Korsgaard's The Sources of Normativity (Cambridge: Cambridge University Press, 1996), chap. 3. It appears here in my discussion as, in effect, a third design specification on autonomous agents.
60. Gibbard, Wise Choices, Apt Feelings, 60.
61. discuss the significance of such cases also in Bratman, "Hierarchy, Circularity, and Double Reduction"
62. and in Bratman, "Two Problems About Human Agency."
63. Gilbert Harman, "Practical Reasoning," in Harman, Change in View (Cambridge, MA: MIT Press, 1986), chap. 8;
64. and Harman, "Desired Desires."
65. See also Alan Donagan, Choice: The Essential Element in Human Action (London: Routledge & Kegan Paul, 1987), 88;
66. John Searle, Intentionality (New York: Cambridge University Press, 1983);
67. and J. David Velleman, Practical Reflection (Princeton, NJ: Princeton University Press, 1989). For an important critique of these ideas,
68. see Alfred Mele, Springs of Action: Understanding Intentional Behavior (New York: Oxford University Press, 1992), chap. 11;
69. Harman's response is in Harman, "Desired Desires."
70. Harman, "Desired Desires," 121. Let me note that I am not here endorsing Harman's general view that all positive intentions are reflexive. I am only using his idea of reflexive intentions to make progress with the special case of self-governed practical reasoning.
71. Ibid., Harman, "Desired Desires", 124. Harman notes here a parallel with John Perry's observation that (as Harman writes) "a child can have the thought that 'it is raining' without having any concepts of places or times and without any inner mental representations of particular places and times, even though the content of the child's thought concerns rain at a particular place and a particular time."
72. Related ideas about reflexivity can be found in Keith Lehrer, Self-Trust: A Study of Reason, Knowledge, and Autonomy (New York: Oxford University Press, 1997), 100-102.
73. I offer a related argument for seeing such policies as reflexive in Bratman, "Two Problems About Human Agency," 323.
74. And, if so, whether the idea of a quasi-policy can usefully capture these similarities.
75. Frankfurt, "Freedom of the Will and the Concept of a Person," 16.
76. But see the modification of this idea in Harry G. Frankfurt, "Identification and Wholeheartedness," in Frankfurt, The Importance of What We Care About, 159-76.
77. See John Martin Fischer and Mark Ravizza, Responsibility and Control: A Theory of Moral Responsibility (New York: Cambridge University Press, 1998);
78. and Keith Lehrer, "Reason and Autonomy" (in this volume).
79. See also Bratman, "Fischer and Ravizza on Moral Responsibility and History," Philosophy and Phenomenological Research 61, no. 2 (2000): 453-58. Note, though, that the present issue is autonomy, not the related but different idea of moral responsibility.
80. (See Gary Watson's distinction between "attributability" and "accountability" in Watson, "Two Faces of Responsibility.")
81. Of course, if the specification of the content of the relevant attitudes is ineluctably historical (for reasons developed by, among others, Tyler Burge and Hilary Putnam), then we would need to appeal to such content-fixing historical considerations. But that is a different matter.