Abstract
Christian List and Philip Pettit develop an account of group agency which is based on a functional understanding of agency. They claim that understanding organizations such as commercial corporations, governments, political parties, churches, universities as group agents helps us to a better understanding of the normative status and working of those organizations. List and Pettit, however, fail to provide a unified account of group agency since they do not show how the functional side of agency and the normative side of agency are connected. My claim is that a constitutive account of agency helps us to a proper understanding of group agency since it ties the functional part of acting to the group agent’s self-understanding and its commitment to specific norms, principles and values. A constitutive model of agency meets much better what List and Pettit seek to accomplish, namely conceiving of group agents as artificial persons, constituted by normative principles and entertaining normative relations to others to whom they are accountable.
This is a preview of subscription content, log in via an institution to check access.
Similar content being viewed by others
Notes
List and Pettit describe a representation and a motivation as a pair of two things, namely an intentional attitude and an object towards which the attitude is held.
The functional model of agency amounts to a more general description of what is commonly called the ‚belief-desire model of agency’. According to that model, every action results from an agent’s beliefs and desires. The agent deliberates from the recognition of having a desire for x, and a belief about the means by which x can be (best) realized, to action (i.e. taking the means for achieving x). The scope of the functional model in List’s and Pettit’s framework is broader than the belief-desire regime. A robot, as List and Pettit claim, can meet the requirements of a functional account of agency (the way the robot is programmed constitutes its transformation procedure and specifies the way in which the robot turns its ‘motivational states’ into movements). In this paper I argue that neither the functionalist model of agency nor the belief-desire model give us the full picture of agency. But I do not draw into question that the belief-desire regime depicts the causal mechanism of acting. However, as I see it, a functionalist account of the robot explains the movements of the robot, but the robot is not an agent in the full sense. We cannot—in a non-metaphorical sense—attribute beliefs and desires to the robot.
In their assessment of majority voting as a model of aggregating individual attitudes into a group agent attitude List and Pettit assume that representational attitudes are judgments, i.e. that p is the case, and motivational attitudes are preferences, expressing a desire that p be the case.
The example by which List and Pettit illustrate the doctrinal paradox is a breach-of-contract case. A court of three judges has to come to a verdict by majority voting on the following propositions: premise 1: The defendant was contractually obliged not to do to a certain action; premise 2: The defendant did that action; conclusion: The defendant is liable for a breach-of-contract. The legal doctrine states that obligation and action are jointly necessary and sufficient for the verdict that the defendant is liable, i.e. the conclusion is true if and only if the premises are true.
The inconsistency by majority voting arises if the following truth values hold for the individual attitudes of the judges:
P1
P2
C
Obligation not to do x?
Action x?
Liable for breach-of-contract?
Judge 1
T
T
T
Judge 2
T
F
F
Judge 3
F
T
F
Majority
T
T
F
There is a majority vote for the truth of both premises, hence the conclusion that the defendant is liable should be true as well. However, the majority vote on the individual conclusions generates the result that the defendant is not liable.
The panel consisting of three experts seeks to form a collective judgment on the following individually held propositions (the propositions p, q in List’s and Pettit’s example are here slightly shortened):
(p): Global carbon dioxide emissions from fossil fuels are above a certain threshold t.
(If p then q): If global carbon dioxide emissions are above threshold t, then the global temperature will increase at least by degree c over the next three decades.
(q): The global temperature will increase at least by degree c over the next three decades.
An inconsistency by majority voting arises if the following truth values hold:
Emissions above t?
If emissions are above it, increase of temperature?
Then increase of temperature?
Individual 1
T
T
T
Individual 2
T
F
F
Individual 3
F
T
F
Majority
T
T
F
A majority of the experts agree on proposition 1 and 2 to be true, yet a majority of experts consider proposition 3 to be false. List and Pettit (2011, p. 45).
The paradox can also arise in the case of preference aggregation by majority voting. See List and Pettit (2011, p. 46).
Universal domain states that the aggregation function allows as input any possible individual profiles of attitudes, assuming that individual attitudes are consistent and complete. Collective rationality ensures that the aggregation function leads to consistent and complete group attitudes. Anonymity secures that all individual attitudes do have equal weight in determining the group attitudes. And systematicity means that the group attitude on each proposition depends only on the individuals’ attitudes towards it, not on attitudes towards other propositions. So “the pattern of dependence between individual and collective attitudes is the same for all propositions” (List and Pettit 2011, p. 49).
List and Pettit’s impossibility theorem which holds for attitudes (namely judgments on beliefs and preferences) is similar to Arrow’s impossibility theorem; the latter shows that any method of preference aggregation which fulfills plausible axioms fails to generate a consistent ranking of collective preferences.
For a detailed discussion why giving up conditions like anonymity and universal domain seems problematic see List and Pettit (2011, pp. 52–54). Giving up rationality is obviously no option.
Constitutive accounts of agency have been criticized in this respect by David Enoch (Enoch 2011). Enoch draws into question that Korsgaard’s and Velleman’s constitutive standards provide foundations for normativity. As I argue in this article I share that scepticism in respect to Korsgaard’s attempt to ground morality directly in agency; but I think that the objection does not undermine Velleman’s account.
In an earlier paper (Enoch 2006), Enoch challenged constitutive accounts of agency more radically by asking: Why should an agent not reject the constitutive aim of agency by simply claiming to be an agent in a different sense, namely a “shmagent”? This challenge is kind of rhetorical because Enoch does not depict the conditions holding for a “shmagent”. For Velleman’s answer to Enoch’s 2006 paper see Velleman (2009, pp. 142–144, and p. 204). Enoch also attributes to Velleman the claim that his account of agency solves the internalism–externalism dispute. But Velleman does not argue for that.
In the 2008 afterword to that article which appeared first in 1997, Korsgaard modifies this thesis; her more recent claim is that the only normative principle of practical reason is the categorical imperative. Her argument is motivated by her downplaying the normative force of the principle of instrumental rationality, which is what Kant means by a ‘hypothetical imperative’. See also footnote 15 of this paper.
The main problem is how the account copes with the problem of bad action. Autonomous agents simply need not be moral persons; they are still agents even if they act in a morally bad or evil way. But a constitutive account of agency need not necessarily be a moralizing account of agency.
Velleman’s views have undergone a certain development. In his article “The Possibility of Practical Reason” (2000b) he held that autonomy is the constitutive aim of action, but he came to revise that position, arguing that intelligibility is the constitutive aim of action. The reason for this revision is exactly the one outlined above, namely that identifying autonomy with taking a critical stance is not sufficiently specific. The idea is that aiming for intelligibility sets restrictions on an individual’s behavior. Certain ways of acting do not make sense, given the aim of an overall coherence in one’s way of leading a life and in one’s self-understanding.
Note that the requirement of instrumental rationality is here formed in terms of a wide-scope requirement, i.e. it can be fulfilled by either giving up the end or by taking the means.
In some of his articles Pettit seems to subscribe to that thesis. He claims, for example, that the belief-desire model requires of me to deliberate my “way to action” not by “mechanically forming and acting on beliefs and desires”, but by deliberating whether a desire or preference gives me a “deliberative reason” to perform a particular action (Pettit 2007a, pp. 40–41). In the same article Pettit concedes that the belief-desire model does not amount to a “comprehensive depiction of agency” (Pettit 2007a, p. 46). Note: Pettit’s argument in this article proceeds via a discussion of the rational choice paradigm. Pettit objects that the rational choice paradigm offers only a partial account of rationality. Since he also holds that the goal-maximizing rational choice schema is equivalent to the belief-desire regime of folk psychology, he thereby admits that the belief-desire model is at best only a partial account of agency. See also Pettit (2002).
Some philosophers have argued that the normative force of the instrumental principle is merely derivative, it depends on the normative force of the ends. See, for example, Korsgaard (2008). Joseph Raz has argued that the instrumental principle is merely a “facilitative principle” enabling someone to function properly as an agent. If one is instrumentally irrational, one is, as Raz claims, “not functioning properly” (Raz 2005, p. 18).
A similar claim is made by Carol Rovane when she argues that agency is tied to a practical commitment to reach overall rational unity (also in a commitment to unifying projects). See Rovane (1988, esp. pp. 179–182).
A famous case is the Chicago-based accounting firm of Arthur Andersen which in 1989 split into two separate branches, an accounting firm and a consulting firm which then harshly competed with each other. In 2000 the two companies split, and the accounting branch set up itself a consulting wing. The firm was then involved in the Enron scandal which finally ruined its reputation, but the dispute between the two branches had already weakened the company. Here we can say that the actions of the group agent simply do not make sense given what the group agent tries to accomplish, namely being an economically successful company.
The list of examples of institutional incoherences, i.e. practices which do not make sense given the constitutional commitments of the group agent, can be extended: political parties showing a remarkable mismatch between the values and ends shaping their party program and the actual behavior of their leading political representatives; churches committed to high ideals of respecting others and (sometimes implicitly tolerated) sexual abuse of minors by priests; trade union leaders helping themselves to high salaries and a lifestyle far from the life of those workers whose interests they should represent.
One might object that the standard of intelligibility is merely relevant to the internal self-assessment of a group agent. Crucial, however, one might claim, is to test the practices of a group agent in light of publicly shared normative standards, such as, for example, rules of ethical economic behaviour. But this criticism rests on a misunderstanding. Ethical standards have to be impartial in the sense that they do not merely hold for some particular group agents, but not for others. But this does not imply that ethical standards have to be external standards, lacking any connection to the standard of intelligibility.
In an overview of his work, Pettit defends a normatively reduced account of agency, namely a so-called deflationary understanding of minded agency. Accordingly, “(t)he agent will act for the realization of certain goals, according to evidentially tuned representations as to how things stand at any moment. It will be a system of intentional states like desire and belief, as it is more often put, where desires select goals, and beliefs constitute evidentially sensitive representations” (Pettit 2007b, p. 229). Pettit assumes that our being rational, i.e. capable of reasoning, involves higher-order or meta-propositional beliefs such as ‘p’ and ‘if p then q’ entails ‘q’. But he is skeptical to include standards of rationality other than consistency and closure. Pettit (2007b, pp. 233–238).
Compared with that account, his common book with List on Group Agency shows a much stronger awareness of the normative elements of agency, though the full implications for an account of agency are not yet taken into account.
List and Pettit call this the performative model of agency, connected to the conception of the legal person in the Roman law tradition where a person is defined by having legal rights and duties, for example, the right for owning and selling property, for making contracts and also for defending their interests in court.
In a reply to Kornhauser and Sager (2004) List and Pettit tend to a personification of groups by arguing that the discursive dilemma poses an integrity challenge for a group, namely holding inconsistent attitudes. This integrity challenge arises when a group treats its members not merely as bearers of preferences or as mere “instruments for testing the truth of some proposition”, but when the group aims at being “an entity that has to answer for its judgements in the manner of a person” (List and Pettit 2005, p. 387). Then the group must be sensitive to inconsistencies. List and Pettit add that this personification needs self-regulation on the side of the group. My point is that this “personification” is exactly the question which norms and standards are constitutive for the group agent and its self-regulation.
References
Enoch, D. (2006). Agency, shmagency: Why normativity won’t come from what is constitutive of action. Philosophical Review, 115, 169–198.
Enoch, D. (2011). Shmagency revisited. In M. Brady (Ed.), New waves in metaethics (pp. 208–233). Basingstoke: Palgrave Macmillan.
Kornhauser, L. A., & Sager, L. G. (2004). The many as one: Integrity and group choice in paradoxical cases. Philosophy and Public Affairs, 32(3), 249–276.
Korsgaard, C. M. (1996). The sources of normativity. Cambridge: Cambridge University Press.
Korsgaard, C. M. (2007). Autonomy and the second person within: A commentary on Stephen Darwall’s The Second-Person standpoint. Ethics, 118(1), 8–23.
Korsgaard, C. M. (2008). The normativity of instrumental reason. In C. M. Korsgaard (Ed.), The constitution of agency. Essays on practical reason and moral psychology (pp. 27–68). Oxford: Oxford University Press.
Korsgaard, C. M. (2009). Self-constitution. Agency, identity, and integrity. Oxford: Oxford University Press.
List, C., & Pettit, P. (2005). On the many as one: A reply to Kornhauser and Sager. Philosophy and Public Affairs, 33(4), 377–390.
List, C., & Pettit, P. (2011). Group agency. The possibility, design, and status of corporate agents. Oxford: Oxford University Press.
Pettit, P. (2002). Decision theory and folk psychology. In P. Pettit (Ed.), Rules, reasons, and norms. Selected essays (pp. 192–221). Oxford: Clarendon Press.
Pettit, P. (2007a). Construing Sen on commitment. In F. Peter & H. B. Schmid (Eds.), Rationality and commitment (pp. 28–48). Oxford: Oxford University Press.
Pettit, P. (2007b). Joining the dots. In G. Brennan, R. Goodin, F. Jackson, & M. Smith (Eds.), Common minds. Themes from the philosophy of Philip Pettit (pp. 215–344). Oxford: Clarendon Press.
Raz, J. (2005). The myth of instrumental rationality. The Journal of Ethics and Social Philosophy, 1(1), 2–28.
Rovane, C. (1988). The bounds of agency. An essay in revisionary metaphysics. Princeton: Princeton University Press 1988.
Velleman, D. J. (2000a). What happens when someone acts? In D. Velleman (Ed.), The possibility of practical reason (pp. 123–143). Oxford: Oxford University Press.
Velleman, D. J. (2000b). The possibility of practical reason. In D. Velleman (Ed.), The possibility of practical reason (pp. 170–199). Oxford: Oxford University Press.
Velleman, D. J. (2009). How we get along. Cambridge: Cambridge University Press.
Acknowledgments
For critical comments on an earlier version of this paper I would like to thank Martin Kusch, David Velleman and the members of the ERC-project “Distortions of Normativity” discussion group, Alexandra Couto, Julian Fink, Christoph Hanisch and Veli Mitova. Research for this paper was funded by the ERC Advanced Research Grant “Distortions of Normativity”.
Author information
Authors and Affiliations
Corresponding author
Rights and permissions
About this article
Cite this article
Pauer-Studer, H. A Constitutive Account of Group Agency. Erkenn 79 (Suppl 9), 1623–1639 (2014). https://doi.org/10.1007/s10670-014-9632-y
Received:
Accepted:
Published:
Issue Date:
DOI: https://doi.org/10.1007/s10670-014-9632-y