I nearly fell. Stumbling and searching for balance, I needed both hands: one for grabbing the doorframe in support, the other for counter balance. My shoulder bag hit my left knee. As it dangled above the ground, I slowly regained composure. Gathering my senses, the first thing that I noticed was a caption printed in large letters that read: ‘Thought Experiment Index’. It was printed on the back pages of the book I had let go of whilst stumbling. My freshly arrived copy of Meghan Sullivan’s Time Biases had hit the ground on the outer end of its spine, propping open its last page to reveal an overview of the thought experiments contained in it. What had happened? – You see, when I get mailed a package with a book, I open it there and then, downstairs, at the mailbox. I will have finished reading the back cover waiting for the lift. And I am usually deep into the Introduction by the time I am at the front door of my apartment. On this occasion, I did not pay enough attention to where I was going whilst stepping out of the lift, and hence the tumble. Afterwards, during my numerous attempts to come to terms with the book, I could not stop thinking about the fact that I had literally stumbled upon the fact that this book has a ‘Thought Experiment Index’. In hindsight, it perfectly foreshadowed that it is a much more exclusively philosophical book than suggested by the publisher.

Time Biases by Meghan Sullivan is a thorough philosophical treatment of issues in which rationality and time interplay, such as personal planning, preference changes, dynamic decisions, and personal deliberation about the future and the past. As the subtitle of the book suggests, it offers a ‘theory of rational planning and personal persistence’. This theory, in its essence, is a rejection of any ‘time biases’, arguing for temporal neutrality all around. The book’s greatest appeal is that in doing so it is unusually comprehensive. Most strikingly and refreshingly, Sullivan talks almost as much about the past as the future: a welcome change to what is a bias towards talking about the future in most of the rationality and time literature. The book will be of great value and interest to any reader in philosophy who already is or wishes to be steeped in philosophical debates about rationality and time. It also makes it a perfect first book to pick up for anyone who is just entering personal identity rationality and metaphysics, perhaps feeling intimidated by tomes like Reasons and Persons (Parfit Reference Parfit1984): it goes through most of the standard arguments in the literature in breezy fashion, covering a lot of ground. And it does more than that. It offers a thorough development and defence of the ‘doctrine of temporal neutrality’.

The doctrine of temporal neutrality unfolds in three steps: Chapters 1–4 present arguments against biases towards the present over the future. They mostly summarize well-known arguments in the philosophical literature for temporal neutrality, but do so in novel ways – and are surprisingly succinct to boot. Sullivan first shows that systematically preferring the present over the future would go against the prudential planning of a well-lived life (the ‘life-saving argument’, Chapter 2). Then she shows that it would be arbitrary to do so (the ‘arbitrariness argument’, Chapter 3 and 4). On that basis, she also engages with the various arguments in the literature about variations in personality providing a possible basis for present bias (Chapter 4). Chapters 5–7 show how similar arguments work for biases against the past, arguing for past neutrality by extending the arbitrariness argument from earlier. Chapters 8–11 develop prescriptive implications of the temporal neutrality doctrine defended. Sullivan maintains that these implications go beyond a general theory of prudential rationality and offer guidance in a great variety of circumstances.

How does Sullivan develop these arguments? Not surprisingly, thought experiments play a major role. The book is indeed peppered with them. Methodologically speaking, that is what is driving the book. Everyone who enjoys reasoning by thought experiment will have the most productive of times with it. To them, it will be especially nice to see the variety of style in thought experiments employed – from the very simple to the quite complex ones. I think I felt challenged getting enjoyment out of this partly because I was – as a philosopher who likes his science – getting hung up on the fact that the sciences were not part of the main show. Indeed, it took me by surprise how philosophical the book was in its conception, argument and tone – given that the publisher advertised it as drawing ‘substantially from work in social psychology, economics and the history of philosophy’. I do not think it does. While many chapters start with examples of empirical research in psychology and economics, little actual use is made of their insights; they are not engaged with on their own merit. Throughout the whole book, philosophy is firmly in the driving seat. A few perfunctory nods are given to the most famous and basic insights of behavioural economics (for example, we are given brief summaries of the ‘endowment effect’ or ‘hyperbolic discounting’). However, nowhere in the book does any content from psychology or economics serve as more than a cue to develop the arguments. And the arguments themselves are mainly presented so as to make them relevant for philosophers of rationality and time. As a result, this book is chiefly of interest for and aimed at the latter group. Since this is quite a strong categorical statement, and since there are some aspects of the book regarding which I am sure the author might want to argue that meaningful interdisciplinary connections are being made, let me illustrate this point.

There is a passage about the psychology and economics of discounting at the beginning of Chapter 1 which does not quite reflect the breadth of current research on the topic in these disciplines. Indeed, in section 1.2 ‘The Received Wisdom’, one specific and contested strand of the psychology literature on discounting (on ego depletion) is at least implicitly suggested to stand for the much larger and sophisticated field of discounting research in psychology and (behavioural) economics. While that may not be a problem in and of itself, there are some missed opportunities to engage with other research on discounting in the later chapters.

Chapter 3 starts with a passage about the ‘future you’. Here, Sullivan cites a well-known type of marketing campaign for stimulating saving and investing in one’s pension – campaigns that aim at making individuals consider their future selves by making these future selves salient (visually) – before going on to develop arguments for temporal neutrality. And later, in Chapter 4, there is a brief discussion of multiple-self models. In both instances, there is no engagement with the literature in psychology, economics or related disciplines when there would be much to gain from it. For instance, the seminal review paper by Frederick et al. (Reference Frederick, Loewenstein and O’Donoghue2002) published in the Journal of Economic Literature makes clear that there are a variety of interesting connections between the (reductionist) literature on personal identity in philosophy and the interpretation of theoretical discounting models in economics. More recently, Heilmann (Reference Heilmann, McIntyre and Rosenberg2016) and Grayot (Reference Grayot2019) showed that there is much need for conceptual clarification regarding the popular use of ‘dual-self’, ‘multi-self’, and ‘multiple-self’ as well as ‘dual-systems/process’ models in behavioural economics. In addition, there is recent work that relates specific notions of these literatures in philosophy to experimental work. For instance, Bartels and Rips (Reference Bartels and Rips2010) and Bartels and Urminsky (Reference Bartels and Urminsky2011, Reference Bartels and Urminsky2015) investigate in how far the perception of connectedness between the subject’s present and future selves coheres with discounting behaviour. At the very least, the main message of Sullivan’s book about investigating temporal biases about the past as much as about the future is an interesting challenge to this literature. At the same time, these references already suggest that the experimental and theoretical literature in psychology and economics are more advanced and sophisticated than Sullivan makes them out to be.

With that in mind, Sullivan’s remark that near bias and future bias should be studied ‘in a more unified and systematic way by social scientists’ (120, Chapter 8) seems a bit puzzling. Surely, with no detailed engagement in any of the literatures just mentioned there are no good grounds on which to make such a demand. I admit that was my first reaction. A more charitable interpretation is that Sullivan makes this remark in relation to the more specific concern of treating near and future bias similarly, as preferences for the scheduling of events, and that those should be investigated systematically. I also fail to see how this claim was argued for explicitly in the book, though. Moreover, I would also say that while some branches of psychological and behavioural economics discounting research have recently focused on specific domains of discounting (such as problems like savings behaviour), the systematic study of discounting is alive and well in economics. The framework of trying to derive and test specific discounting functions is one of systematically studying scheduling preferences. So, economic theories of discounting – theoretical, experimental and behavioural – seem to fit the bill for Sullivan. I do however think that Sullivan’s challenge to study preferences for the past more is interesting in relation to this literature as well.

The latter point brings me back to the most tangible achievement of this book: having comprehensively expanded the doctrine of temporal neutrality to the past. Demonstrating that is a matter of working through both the metaphysics and rationality of temporal neutrality. Of the merit and success of that project I am convinced. It stands separate from the concerns I raised about engaging with other disciplines and their scientific goals. The interdisciplinary question, though, can also motivate the following, more constructive way to look at the thought experiments in the book: as a way to potentially interest psychologists and social scientists into empirically investigating some questions and scenarios that Sullivan considered. And for that, the ‘Thought Experiment Index’ may come in handy, too.