Geoff Brennan died in Canberra on 20 July 20 2022, aged 77.

The phrase ‘a scholar and a gentleman’ could not be more fittingly applied than to Geoff Brennan. He was a true scholar, happy to discourse at length and in depth over coffee or dinner or in the academic settings of conferences and seminars. He was also a great listener, and took others’ ideas seriously, whether they were famous academics, undergraduates or dinner companions outside of academia. The fact that he listened and rethought his own positions is something that defines his own life’s work. He sought to delve for the truth, and not to simply to defend his own opinion. And he was a true gentleman, generous in spirit and deed. In tributes many have said how positive and encouraging he was to them. He was not shy of expressing criticism, however, though accompanied with a smile and a twinkle and formed so broadly (‘I’ve often wondered why people think …’) that one would hardly realize one’s own views were being critiqued. He leaves behind a substantial body of work and an enormous number of friends and colleagues whose own work and lives were enriched through their interactions with Geoff.

Geoff was devoted to PPE (Politics, Philosophy and Economics). Together we set up the PPE program at the Australian National University (ANU). It tells you something about Geoff that he did not really like how successful the Bachelor of PPE degree became in the eyes of the university. Despite being an elite degree with a higher entry standard than for studying each of its components separately, it attracted hundreds of students. Geoff didn’t really want that. He would sooner have had an intake of 20 or 30 students, so he could know them all individually and personally help them with their passions. He cared about the whole person. At a meeting set up by ANU marketing to advertise the forthcoming PPE program, Geoff was talking to some prospective local Canberran candidates. ‘Oh, don’t come to the ANU’, he said, as the marketing department looked on in horror, ‘university is about getting away from home life and parents, and drinking coffee in your rooms until the early hours of the morning discussing God and the universe. Go to Sydney or Melbourne, don’t stay in Canberra.’

Geoff was a prolific author and wrote across a variety of PPE topics: public goods, taxation, electoral preferences, norms, feasibility, reputation. If one merely dipped into his (often-co-authored) essays, one might think he was a proponent of Albert Hirschman’s predilection for petites idées − little ideas that illuminate and instruct. It would have been a mistake. To be sure, he applied his wealth of ideas across many subjects, often with surprising or counter-intuitive results, but they lay within an overarching framework, developed over time; they are built around the edifice of what he defended as revisionist public choice theory. Geoff said himself, on receiving the Hayek Medal in 2014:

I am by disposition, and increasingly by academic self-identification, a ‘PPE person’. By this I mean that, though my home discipline is Economics, I am inclined to focus on the intersections between Economics and Philosophy, and between Economics and Political Science. (Brennan 2016, p. 331)

Geoff began in public economics, examining issues around equity in tax shares and public goods provision and around Pareto-optimal redistribution. That early work foreshadows some of his ‘revisionist public choice’ ideas by suggesting that income transfers can be a public good if people are mildly benevolent donors (Brennan, 1975), though self-interest also provides reasons for redistribution − to suppress civil unrest (Brennan, 1973). His work on Pareto-optimal redistributions caught the eye of James Buchanan and led to Geoff visiting the Center for Study of Public Choice in Blacksburg and then to a close collaboration with Buchanan, notably in two important books, The Power to Tax (1980) and The Reason of Rules (1985).

Geoff’s work with Buchanan was firmly in the public choice tradition, with the customary public choice assumption that people − including government agents, of course − are self-interested maximizers. The Power to Tax is about constitutionally limiting the scope of government discretion to tax for personal or political advantage. It caused some controversy as the Leviathan implications of a revenue-maximizing government offended academics in the welfare economics tradition that sought to advise a presumably benevolent government on optimal tax policy. The Reason of Rules was, in part, designed to answer critics and spell out the argument that constitutional design is separate from day-to-day political processes. That is, there is nothing wrong with providing government with good reasons to tax or to spend, but one must restrain its abilities to exploit those good reasons for other purposes. Constitutional constraints do not curtail reasonable political activity. In Geoff’s work with Buchanan, constitutional design based on those public choice principles becomes the major task of political economy, fitting nicely with the mainstream Rawlsian approach to political philosophy. Some of Geoff’s later work adopts the same constitutional framework to critique arguments within post-Rawlsian political philosophy.

Constitutional design remained central to Geoff’s work, even as he later departed from some of the behavioral assumptions of mainstream public choice found in those first books. We might also reflect on what led Geoff to abandon those assumptions. To be sure, few who work within the public choice tradition really believe that everyone is so venally self-interested, but the general self-interest assumption is defended on Humean risk-averse principles by treating everyone as knaves, especially when it comes to constitutional design. However, perhaps Geoff realized that the move made constitutionally at the collective level likewise can work at the individual level, recognizing that how one wishes to set up incentives for one’s own behavior is one level removed from one’s day-to-day behavior. How one develops principles to guide one’s actions is different from the immediate reactions one has to any situation. That is, one might try to develop a personal constitution, a dispositional outlook, that provides guidance and personal incentives for our day-to-day decisions. Geoff worked this out most thoroughly in his book with Philip Pettit, The Economy of Esteem (2004), but it infuses all his post-Buchanan work. Having come to that realization, one might wish to develop constitutions that support the development of such principled moral motivations at the individual level, something Geoff explored with Pettit and also with Alan Hamlin.

Geoff started to critique traditional behavioral assumptions when he was still writing with Buchanan. Their work on the differences between voting and market behaviour foreshadowed Geoff’s expressive theory of voting, arguing that while voters participate in elections and care about political outcomes, ‘there is no logical connection between these two facts’ (Brennan & Buchanan, 1984, p. 187). The reason, of course, is that a single vote has a vanishingly small causal effect on who wins and so the cost of voting suggests zero participation. We have to get voters to the polls for non-instrumental reasons. Having said that, voters clearly care about the result.

The expressive theory of voting was worked out in detail with Loren Lomasky in Democracy and Decision (1993). They argue that voting cannot be a purely instrumental act, on the grounds that a single vote really does not make a difference to the outcome; thus, voters do not really choose among social alternatives. That is, rather than choosing between two alternatives x and y, voters are expressing a preference for, say, x over y; they are choosing a vote for x over y. Brennan and Lomasky suggest that voting is like cheering for your favorite football team (a metaphor Geoff often repeated).

Democracy and Decision provides a set of empirical expectations defended by what we observe in electoral activity, and a set of normative implications that are much more controversial. An important empirical implication, which Geoff later followed up in work with Hamlin, is the failure of the median voter theorem, with an expectation of, at least with plurality and alternative vote systems, bimodal distributions. Another implication is that we should see less cycling than cooperative game theory applications predict, since expressive voters reward candidates and parties that display integrity and stability in their views. And that worked-through version of expressive voting reaffirms that people will support redistributive policies even if they do not gain, since it is less costly to support collective welfare by voting than it is by directly providing welfare themselves. In other words, expressiveness provides stronger incentives for action given the ‘mildly benevolent donors’ in Geoff’s early foray into welfare issues.

The most controversial of Democracy and Decision’s normative implications is that the preferences that are revealed at the polling booth do not have the same authenticity as preferences revealed through the market. Revealed preferences in market situations show what people want, given the alternatives available. Voting reveals only what people express as their preferences, not what they actually would choose if their vote determined results. In later writings, Geoff pulled back somewhat from that more radical claim. He accepted what some critics pointed out, that while altruism is cheaper in voting contexts than market ones, that does not make it any less of a preference. Votes surely can reveal how people think society ought to be, even if they would be tempted to cheat if they could determine the actual outcomes. In other words, votes can express the preferences that individuals believe should be developed in society, rather than those preferences that are their entirely self-interested ones. By the same argument, the ‘veil of insignificance’ radically reduces the relevance of narrow interest at the ballot box. So, while voting is a symbolic rather than an instrumental act, it can still symbolize the voter’s considered moral judgements.

In Democratic Devices and Desires (Brennan & Hamlin, 2000), expressive voting allows that elections can be decided based on voters’ moral principles and the ethical appeal and virtues of the candidates. Moreover, such preference revelation is defended as superior on long-run material grounds, in the sense that normative behavior is in the long-run interest of society and thus of individuals within it − even if each individual would be still better off if they could exclusively free-ride. The book works through the further normative implications of the expressive attitude.

Brennan and Hamlin produce an original argument for the justification of representative democracy and its superiority to direct democracy. Expressive reasons suggest that voting directly on policies cannot activate an invisible hand, since individual votes do not reflect material self-interests − the input is of the wrong kind, they say. Nor can we rely on expressive votes reflecting a perceived public interest, since expressive voting is as likely to reveal particular enthusiasms or protests as it is the general will − think of Brexit! Either way, aggregating across sets of votes on direct issues will not produce responsible government. However, if voters choose representatives, then they will reflect on issues of competency, virtue, stability of views. Brennan (1998) suggests that voters might prefer a candidate whose views are known to be stable and secure than one whose views are unstable, even if the latter espouses opinions closer to the voter’s bliss point in policy space. He argues that voters will prefer the certainty provided by the trustworthy candidate to the uncertainty provided by the chancer.

Brennan and Hamlin argue further that the preference for certainty provides a major function of parties. As collective organizations, parties create more stable policies and therefore greater trust in what a government will do than we get from individual candidates alone. It is an argument for the strong party systems we see in parliamentary democracies, rather than the weaker parties in presidential systems, and provides a strong antidote to much of political philosophy, where parties are either ignored or spurned.

Democratic Devices and Desires utilizes the distinction between our blunt preferences, what we want here and now, versus what dispositions we try to inculcate in ourselves. While not directly critical of The Reason of Rules as such, it suggests that, rather than providing a constitution to guard against our worst behavior, we should try to develop one that encourages our better behavior. Although it also makes very clear that we must not be too starry-eyed; we need to restrict the ability of government to prey upon the citizenry.

With Phillip Pettit, Geoff took even further the idea that dispositional motivations are important when public and private esteem are sought. One sets goals for oneself, moral goals, in the belief that there are certain principled ways of behaving, and one wants others to recognize and esteem oneself. In an early article, Brennan & Pettit (1993) argue that an important set of what they call ‘decentralized and contingently non-intentional responses’ leads people to better behavior. People do not take certain actions because they would be embarrassed to be caught doing so, even if they were not sanctioned. We internalize the sanctions that could be imposed − friends not talking to us, or even friends saying nothing but knowing they would think less of us.

The same approach also can work at the political level, with politicians thinking of their legacies, wanting to be thought of as good people. The Economy of Esteem explores those issues in greater detail. It considers the paradoxes of public esteem wherein the very attempt to gain esteem is inconsistent with deserving it, while also recognizing that individuals with weaker private commitments to principled behavior can be motivated to display such principled behavior in order to be esteemed publicly. In that argument the interplay of private and public esteem is important in keeping public actors from the stark predatory behavior that traditional public choice assumes.

The multi-authored book Explaining Norms (Brennan et al., 2013) suggests that norm following is a non-instrumental desire to act in accordance with the norm because one believes that compliance requires such action in a given situation. Brennan et al.’s account of norms is non-reductive, since the normative component of norms is, they argue, irreducible. We cannot explain norms simply by their functioning or optimality. Here again we see explanations of social behavior that are far removed from the standard economic accounts, and certainly from standard self-interested rational actor accounts of social phenomena. However, Explaining Norms does not depart from rational actor assumptions, and Geoff’s influence in the more formal parts of the book is clear.

Indeed, Geoff has spent considerable effort in trying to bring economic thinking into moral and political philosophy.Footnote 1 We can see two important components in his thinking here. The first is to press constitutional thinking into political philosophy, rather than seeing it as a branch of moral philosophy. The second is concerned with feasibility. The two are related. Geoff always was concerned about issues of feasibility when considering desirable and feasible taxation schemes. But his real concern with feasibility is that the standard political philosophy approach (at least in ‘ideal theory’) goes something like this: Let’s think of what is morally desirable, and then work out how we might bring it about. In other words, the feasibility issue is separate from the desirability issue. However, feasibility needs to be built into what we think of as desirable, since feasibility frontiers themselves involve tradeoffs between different values. Put the same thought together with the idea that constitutions − both external and internal − govern our incentives to behave in ways we think appropriate, and we can see why feasibility and desirability are not completely separate. Feasibility always is a matter of degree as is, more standardly thought, desirability. Geoff’s analytic defense of conservatism likewise might be viewed in the light of his thoughts on feasibility (Brennan & Hamlin, 2004).

Geoff’s work was recognized widely in the academy. A Fellow of Australian Academy of Social Science, with an Honorary Doctorate from the University of St Gallen (1987), the Distinguished Fellow Award from the Australian Economics Society (2013), and the Hayek Medal (2014), he was the first non-American to be President of the Public Choice Society. He was Head of Economics at the ANU before becoming Director of the Research School of Social Sciences (1991−1996). As is often the case with ‘rational choice’ scholars, Geoff did not act in the self-interested manner public choice assumes those in power will adopt. Indeed, he remarked not long after leaving that post that until the final months he had not really understood how much power it afforded him.

Geoff latterly held positions at both the ANU and the University of North Carolina, spending part of the year at each institution. As well as co-founding the ANU BPPE program, he was founding director of the Duke/UNC PPE program. He also was instrumental in organizing a ‘Winter School’ at the University of Notre Dame, in Fremantle, Western Australia. As ever, he was championing the PPE approach to students.

Geoff’s academic output was enormous, but I never was sure how he did it. When did he write? He always seemed to have time to chat at length about any subject − a morning coffee could last two hours if you were not careful, a chat in the corridor almost as long. While he often wrote alone, a great deal of Geoff’s work was co-authored and I think he liked to work with others because it forced him to finish things. For Geoff there was always something more to say on a subject. His mind never was at rest with what might appear to others to be the answer. And there were never two sides to a question but multiple dimensions, and he was happy to hear your perspective.

Like many Australians, his relaxed attitude hid a competitive side. He took his golf and his singing very seriously, and I recall one occasion when he returned from a family gathering feeling he had not been in good voice; he was really quite downhearted that he had not performed well and been bested by his brother. His love of wine, red wine, was legendary. He had a fine cellar and was prepared to pay more for a bottle or a case of wine than he was willing to admit, especially to his wife Margaret! He once told me had been well paid for some speaking engagement and Margaret said, ‘you never did tell me how much you were paid − you spent it all on a crate of wine.’ Never had Geoff looked so sheepish.

At Geoff’s funeral, we heard that he did not fear death, but felt joyful for the good life he had lived. Geoff was sustained by his deep Christian beliefs, something over which he once had a serious conflict with Buchanan (recounted in Brennan & Munger 2014). Five years ago, at his retirement dinner, he admitted to me that he did have fears that with increasing age his mind would lose its acuity. It did not. Knowing the poor prognosis for his illness, he worked until the end. In the last few months of his life, despite the fatigue that leukemia and chemotherapy impose, Geoff completed revisions for papers and worked on two books, Ethics and Economics and one with Hartmut Kliemt. On Exchange is an extended justification of Buchanan’s advice that economists should focus on exchange rather than optimization. Rejecting Buchanan’s ideal of ‘politics as exchange’ and also conventional folk theorem logic, exchange-enabling institutions are defended in broadly Smithian terms as the socially and legally embedded engine of the wealth of WEIRD (Western, Educated, Industrialized, Rich, Democratic) nations.

Over the last decade or so, Geoff was involved in an ‘implausible juxtaposition’ as he put it: a ‘PPE and Literature Colloquium’ run at UNC and occasionally at the ANU. At the time of his death, he was looking at Adam Smith and Jane Austen; his final contribution, ‘Smith and the Formation of Attitudes’, mused on a passage in Smith’s Theory of Moral Sentiments, about our empathy even with the dead. Geoff did not accept Smith’s analysis. He writes:

I reckon there is a sharp distinction between the (proper) attitudes of A to her own death and the (proper) attitudes of close survivors B C D etc. A stands at the end of a very happy life. She is intensely grateful to all those who have fallen in with her madcap schemes and enriched her life. When she examines that life as a whole, to feel the slightest tinge of reproach because she did not get a bit more of it seems … well, greedy — somehow churlish. She finds herself principally indifferent between a date of her death in three months or six. To be sure, she hopes to avoid pain in the process. But she finds in herself a certain serenity at the prospect of death itself — and this, as she sees it, is totally unconnected with any thoughts about the afterlife. (Brennan 2022)