Being Good in a World of Need, Larry S. Temkin. Oxford University Press, 2022, 432 pages.

The goal of Being Good in a World of Need is to force readers to question commonly held assumptions about global aid efforts. Larry S. Temkin aims to do this by providing an empirically informed ethical analysis of the topic. Given the empirical complexity of the subject matter, Temkin is careful not to offer a decisive verdict on the ethical status of contributing to global aid efforts. Instead, the central thesis of Being Good is cautionary: contributing to global aid efforts is more ethically troubling than some influential commentators have let on in the past – including Temkin himself.

The timing of Being Good could not be better. The Effective Altruism (EA) movement, a major focus of the book, is currently undergoing a public reckoning due to the collapse of cryptocurrency exchange FTX. ¹ But it would be a mistake to suggest that Being Good is just for adherents of EA. Despite the influence that EA has had on global aid efforts over the past decade and a half, there are still many people unaffiliated with the movement who have an interest in combatting global poverty. Temkin makes this clear in the first chapter and includes philosophers, economists, global aid experts, aid workers and ordinary people concerned with worst-off as part of his intended audience.

The rest of this review is split into two parts. First, I focus on Temkin’s ethical analysis and highlight what the most important contribution of Being Good is. As I suggest below, both philosophers and economists stand to learn valuable lessons from Temkin’s pluralistic approach to global aid. In the second part, I focus on Temkin’s engagement with the empirical literature on global aid. Even though it is refreshing to see a leading moral philosopher engage with empirical literature, this aspect of Being Good does not quite meet the standards Temkin sets for doing empirically informed philosophy. Economists will want to take note, however, as the empirical shortcomings in Temkin’s analysis provide opportunities for important future research.

1. A Pluralistic Approach to Global Aid

The philosophical backbone of Being Good is presented in the second part of Chapter 2. Here, Temkin outlines his pluralistic approach to moral philosophy, which he applies to various dimensions of global aid efforts. Temkin follows contemporary philosophical orthodoxy in dividing ethical theory into three major traditions: consequentialism, virtue ethics and deontology. Each tradition is, in turn, the source of distinct kinds of ethical reasons: outcome-based, virtue-based and deontological-based – all which should enter our deliberations about what we should do.

Temkin’s pluralistic approach to global aid is without a doubt the most important contribution of Being Good. As Temkin points out in Chapter 3, philosophical discussions of global aid have largely been dominated by the consequentialist tradition in moral philosophy and specifically its utilitarian branch. This is in large part due to the influence of the EA movement – Temkin traces its origins back to the utilitarian tradition, but also to some form of welfarist consequentialism being the dominant evaluative framework in mainstream economics.

In Chapter 3, Temkin begins a lengthy argument spanning most of the book. The main conclusion of this argument is that contributing to global aid efforts is more ethically and empirically complex than Singer’s famous Pond Example suggests. For readers unfamiliar with ‘Famine, Affluence, and Morality’, Singer (1972) uses the hypothetical case of saving a drowning child from a shallow pond to provide support for the following moral principle:

Preventing Bad without Moral Loss: If it is in our power to prevent something very bad from happening, without thereby sacrificing anything morally significant, we ought, morally, to do it. (Singer 1972: 231)

Singer then famously uses this principle to argue that people who are reasonably well-off should help those dying from famine or easily preventable diseases by contributing to global aid efforts. Contributing to global aid requires us to give up some disposable income, but this is not a morally significant sacrifice per Singer and his followers.

Chapter 3 showcases why Temkin is considered one of the greatest living moral philosophers. The chapter is notable in how it establishes that virtue-based reasons are relevant to questions about helping those in need. To borrow the book’s title, the key contribution is that ‘being good in a world of need’ consists in more than ‘doing the most good’ (the mantra of the EA movement). Through a series of thought experiments, Temkin elicits our intuitions about moral character to convincingly show that, when it comes to helping those in need, sometimes moral agents are not required to perform an action that is optimal as judged by the lights of consequentialism. One of the most important thought experiments is Sudden Epiphany (67). Temkin asks us to imagine a reasonably well-off person confronted with a drowning child who, at the crucial moment of deciding whether to jump in or not, realizes he could save more lives by not jumping in and selling an expensive non-waterproof watch (which he cannot take off) and then donating the proceeds to an effective aid organization. Since the aid organization will save more lives, he lets the child in front of him drown.

Many readers will accept that letting one child in front of you drown is wrong even if this will lead to more innocent lives saved abroad. But the point of the thought experiment is not simply to pump the reader’s intuitions against consequentialism. The thought experiment leads to a crucial insight: the ethical impulse to rescue people in need in both real and hypothetical scenarios should be disentangled from general duties of beneficence. Those who invoke Singer’s Pond Example have routinely overlooked this point. Temkin argues that direct confrontation with those in need of rescue (such as in Pond Example or Sudden Epiphany) generates a special obligation that can take precedence over the general duty to do good. Temkin still maintains, along with Singer, that duties of beneficence are invariant to geographic distance. ² But the crucial lesson is that the duty to rescue stems from virtue-based reasons, not consequentialist reasoning. As Temkin elegantly puts it, ‘human life would, overall, be sadder, bleaker, colder, less valuable, and less praiseworthy, if we lacked the virtues and spark of humanity that propel us to make costly dramatic rescues despite the inefficiencies of such rescues’ (83).

Temkin does note that Singer’s principle can accommodate such virtue-based concerns since what is ‘morally significant’ does not have to be restricted by consequentialist criteria. But the main point is that these virtue-based considerations have long been overlooked, and Temkin should be given credit for emphasizing that they need to play a more central role in our thinking about how to respond to those in need.

Subsequent chapters also show how deontological considerations are relevant to global aid efforts. Chapters 5–7 deal with ‘the dark side of humanity’ and discuss corruption both within international aid organizations (‘internal corruption’) and among the communities receiving foreign aid (‘external corruption’). One would have to be naive to not know that bad people sometimes operate within organizations that promote noble causes, or that the developing world is marred by brutal autocracies and ethnic violence. The main point is not just that Singer’s Pond Example abstracts these features away. The more important lesson is that there are overlooked deontological worries that donating to an international aid organization raise. As Temkin points out, getting aid to those in need may result in harm to innocent individuals and benefits to malevolent actors. Temkin is not suggesting there should be absolute deontological constraints on ‘dealing with the devil’. The contribution is that there are morally significant deontological factors to consider when contributing to global aid efforts. ³

2. Engagement with Empirical Literature

Much of the rest of Being Good draws from the chapter on foreign aid in Deaton’s (2013) The Great Escape. Temkin notes early in the book that a potential weakness of Being Good is that it relies too much on Deaton’s views. But as Temkin points out, Deaton is a respected Nobel Prize-winning economist. Temkin is faultless in suggesting we take Deaton seriously. Yet, there is a missed opportunity in how Temkin frames Deaton’s concerns and more generally how he engages with the empirical literature on global aid. There are many signs throughout the book that Temkin is well-aware of the criticisms I will raise, and my goal here is only to hold Temkin to the standard that he sets out for himself as a philosopher:

My job, as a philosopher, is to help identify some of the important empirical and normative issues that are relevant to how we should respond to the needy. But as for gathering and assessing any relevant empirical data, I must leave that to the economists, sociologists, political scientists, anthropologists, and other social scientists, as well to aid agencies, aid evaluators, and governments. (338)

There is nothing wrong with the division of labour that Temkin is proposing. The problem is that, while the empirical literature Temkin cites is important, its relevance to the question of how individuals should respond to those in need is not as straightforward as Being Good suggests. MacAskill (2019) has raised a similar criticism in response to earlier work which Being Good is based on. My concerns echo MacAskill’s, though my focus is on what claims Deaton’s published work support.

The last third of Being Good takes up a consequentialist perspective and concerns what Temkin aptly dubs ‘Deaton’s Worry’. In a nutshell, Deaton’s Worry is that global aid efforts are not only ineffective but are in fact doing more harm than good. The basis for this controversial claim is a plausible explanation for a phenomenon that development economists call the resource curse: Why have so many low-income yet resource-rich countries fared so poorly in recent history? Deaton’s preferred explanation is that easy access to natural resources crowds out a government’s incentive to finance expenditures by taxing their citizenry. This, in turn, leads to poor institutions and lack of state capacity since governments have no incentive to become accountable to their citizens by taxing them. Deaton argues a similar mechanism operates when foreign aid is introduced in developing countries (an ‘aid curse’). A flow of foreign aid has similar effects to a resource boom in a developing country, i.e. it hampers economic development by making governments unaccountable to their citizens. For reasons made clear below, understanding the causal mechanism behind Deaton’s Worry is important.

Temkin takes Deaton’s Worry to suggest that altruists face a special kind of collective action problem. Temkin calls this a ‘consequentialist each/we dilemma’. ⁴ Temkin starts by accepting that each one of us has an outcome-based moral reason to help someone in need by contributing to global aid efforts. However, the aggregate effect of everyone acting on their individual moral reason to promote good outcomes is that we cause governments in the developing world to become unaccountable to their citizens. This ensures that in the long run the poor remain poor and in need of our assistance. Our collective aid efforts thus result in more harm than good by creating an impediment to institution building in the developing world – a rather depressing conclusion, if true.

The EA movement face a damning criticism of its prescriptions if Temkin’s analysis of Deaton’s Worry has even some chance of being correct. But while the concept of a consequentialist each/we dilemma is theoretically interesting, the empirical basis for Temkin’s application of the concept is not as strong as the book suggests. The problem stems from Temkin not making salient some important nuances about the global aid literature. First, there is a distinction to be made between official aid and unofficial aid. Official aid is either bilateral – meaning it is given from government to government – or multilateral – meaning it is allocated from government to government via a multilateral organization such as the World Bank. Unofficial aid is the aid that comes from international charities and other NGOs operating in developing countries. The kind of aid Temkin is referring to throughout the majority of Being Good is unofficial aid, which accounts for 25–30% of transfers from rich countries to poor countries (Deaton 2013: 176). Crucially, this is the kind of aid the EA movement has advocated for and which the movement has contributed most to.

Not taking this distinction seriously leads to complications. Much of the literature expressing scepticism about aid effectiveness is concerned with official aid. Easterly’s (2006) well-known book on the subject, The White Man’s Burden, is a notable example. ⁵ But to be fair to Temkin, it is important to focus on what Deaton’s published work suggests since, as already acknowledged, Deaton is worth taking seriously. However, much of the discussion of foreign aid in The Great Escape is also concerned with official aid. So, there is already some basis to question how much bearing Deaton’s claims have on the EA movement’s efforts despite Deaton’s rhetoric in The Great Escape.

Temkin acknowledges this criticism in parts of Being Good. In a footnote (310), Temkin suggests that most development economists would accept that the distinction between official and unofficial aid is a difference of degree and not kind, so the problems bearing on the former apply to the latter and vice versa. But Temkin does not cite any literature in support of this claim. Temkin instead suggests the claim is supported by his survey of the empirical literature – a departure from the division of labour he proposes. In The Great Escape, Deaton does address the matter in passing and suggests that since aid is fungible, ‘schools and clinics operated by NGOs may [my emphasis] free up funds for the government – and governments find ways of taxing (or simply diverting) the NGOs’ resources’ (Deaton 2013: 279). This is certainly possible but hardly decisive. There is an important opportunity for further research here. ⁶

A second subtlety that Temkin is aware of, but does not make salient, is that much of the aid literature is focused on one specific development goal: economic growth. This is Deaton’s focus in The Great Escape with one notable exception: public health. Deaton singles out aid-funded health projects in developing countries as an exception to his claim that foreign aid does more harm than good. Deaton has made his position clear: ‘External aid has saved millions of lives’ (Deaton 2013: 307). This is important since the EA movement’s aid efforts have centred on micro-level health interventions such as increasing access to preventative health products (e.g. anti-malarial bed nets) – not on spurring economic growth. Even granting there is a causal pathway from official aid to poor institutional quality, it is not obvious that the same mechanism operates through unofficial aid efforts directed at public health. It is still possible that ‘vertical’ aid-funded health projects interfere with the development of the state capacity needed for ‘horizontal’ primary health system to operate successfully – something Deaton does discuss in The Great Escape. But again, this possibility is only mentioned briefly in Deaton’s discussion of foreign aid, and ultimately Deaton (2013: 311) suggests that there still may be an important role for foreign aid to play in funding certain health projects in developing countries.

At various points in Being Good, Temkin emphasizes many of Deaton’s qualifications about global aid efforts; this suggests he is well-aware of the points I am raising. But Temkin’s framing of Deaton’s Worry is nevertheless misleading. Instead of asking ‘What if Deaton is right?’ – a question Temkin is preoccupied with in the last third of the book – we should ask ‘When is Deaton right?’. In other words, when will contributing to global aid efforts undermine local institutions and erode state capacity? ⁷ The latter question is better supported by the empirical literature and still leaves room for the ambitions of the EA movement – at least the segment concerned with helping the global poor. And at times, Temkin does seem to frame the discussion around the latter question when he provides practical guidance on how to do good going forward.

In the end, Being Good does not provide the devastating critique of the EA movement some may be seeking. But as highlighted earlier, adopting a pluralistic approach to global aid is of great importance, and so is being sensitive to the role institutions play in promoting development goals. Being Good should be commended for calling attention to these themes.