My comments on Kahan’s new history of liberalism have a relatively narrow focus: the book’s organizing concept, fear, rather than its insightful accounts of different authors and movements. I suppose that makes me a “lumper,” in the familiar dichotomy between lumpers and splitters that Kahan cites in his introductory chapter. But I’m afraid that even historians have to lump it sometimes. For you need to know just what idea you’re splitting up, even if your goal is to trace its different manifestations. You need to identify the genus correctly if you want to record the lives of its most interesting species.

Kahan recasts the history of liberalism as a series of waves crashing against different objects of fear. He takes his cue from Judith Shklar’s influential essay on the “liberalism of fear.” Like Shklar, he uses the idea of a liberalism focused on diminishing fear as a means of correcting overly abstract and unpolitical accounts of liberalism and its history. This results, among other things, in a history of liberalism in which contractualism recedes to the background, except for a brief appearance with Kant and a late encore with Rawls. After teaching a couple of generations of students to identify liberalism with Rawls’s contractualism, we really need such a correction. Kahan agrees with Shklar that liberal politics pursues a negative goal, diminishing the kind of fear that makes it difficult, if not impossible, to act freely. But he recognizes that Shklar’s conceptualization of the liberalism of fear is too narrow to serve as the framework for a comprehensive history of liberalism. It can only serve this purpose if broadened beyond her relatively narrow focus on our fear of cruelty.

How should we do that? Here’s where things become a little unclear. We could simply multiply the objects of fear that attract the attention of liberals at different times and places. Shklar’s mistake, from this point of view, was to focus on too specific an understanding of the genus of things, fear, that liberalism sets out to counteract. In line with that approach, Kahan constructs a short list of the objects of fear that have preoccupied liberals: religious fanaticism, revolution and reaction, poverty, totalitarianism, and populism. It’s a good list, one that sets lots of things right and promotes lots of new insights. But if the problem with Shklar’s argument is her confusion of a species of liberal fear with its genus, then we need to broaden our understanding of the genus, not just identify its different species. Kahan seems to recognize that as well. He notes in his first chapter that you can’t simply identify liberalism with an effort to eliminate fear. It must be a more focused object of fear. Fear of mortality or accident or natural disaster doesn’t cut it. But I don’t think he gets much farther in specifying the genus of fear whose species he explores so vividly in the book.

Why is it important to do so? It certainly was not very important for Shklar, who is one of the least lumpy political theorists to enter the canon. But then Shklar was not interested in surveying the history of liberalism. One reason for clearing up this genus-species problem is to help us identify liberalism as an object of study, approbation, and critique. Talk about liberalism as the struggle for a world in which we have nothing to fear, and it begins to sound like the savage caricature of progressive politics that Nietzsche drew so memorably in Beyond Good and Evil.

Liberalism, from this point of view, looks like an effort to make the world safe for Nietzsche’s “last man,” the individual whose notion of the good life is threatened by fear of opposition, discomfort, and indigestion. Of course, to the extent that liberals have allied themselves with modern science and technology to find ways of easing the discomforts of existence, they cannot ignore such criticism. But it is precisely because they devote themselves to attacking a more specific object of fear, one that focuses on the ways in which human beings exercise power over each other, that they can answer critics like Nietzsche and point to the many ways of diminishing suffering that they will resist—like genetic engineering—because these put too much of the wrong kind of power in some people’s hands.

So I would suggest that any broadening of the genus of the liberalism of fear should maintain some of Shklar’s focus on the exercise of power. A good place to start building a better lump of liberal fear would be Montesquieu’s account of freedom as rooted in our sense of our security, which Kahan discusses with great intelligence. Unfortunately, doing so makes it very hard to include fear of poverty as one of the species of liberal fear, as Kahan insists on doing. It’s easy to see what is shared by Kahan’s other liberal waves, fear of religious fanaticism, revolution/reaction, totalitarianism, and populism. Among other things, each focuses on something that expands and deepens the power that we exercise over each other. Identifying them as members of a common genus is thus a relatively easy call. But fear of poverty is another matter. Like Roosevelt’s freedom from want, it invokes the fear of being unable to satisfy natural needs. Fear of poverty, no doubt, can undermine one’s freedom, as anyone who has literally or figuratively slaved at a job will certainly attest. But is it a distinctly liberal object of fear? (Which is not the same question as asking whether liberals could or should try to minimize it.) Fear of poverty is harder to fit into Kahan’s story because it focuses less on freedom from a certain kind of subordination than escaping a more generalized form of insecurity.

Indeed, it’s not until you come to John Rawls and his “difference” principle that an argument specifically modeled on fear of poverty gains a prominent role in liberal theory. Here was a theory that made addressing our insecurity about the resources needed to live one of the two basic principles of liberal justice. Fear of getting the short end of the stick, Rawls argues, makes it rational for his social contractors to maximize their minimal resources, which means choosing a principle of justice that justifies inequality only when it promotes the good of the least advantaged. But because that identification of rationality with the so-called “maximin” principle is hard to justify, the initial enthusiasm for Rawls’s synthesis of liberalism with welfare state progressivism soon died away. The Rawls industry, which is still operating at full employment levels, shows little interest in the difference principle nowadays.

Not that the arguments of libertarians who demand that we treat state efforts to eliminate poverty as a new object of liberal fear fare much better. Kahan does a nice job of showing how libertarians import their fear of despotic power into their image of state regulation. That fear makes some sense when you are talking about Soviet-style central planning, as most progressives would now agree. But in order to employ it against run of the mill welfare state measures, like national health insurance, libertarians have to play a game of bait and switch in which they hope no one notices that they are applying arguments about central planning to distinctly mixed economies.

In the end, it is important to remember that liberalism is a partial ideology, as Kahan himself notes. It does not pretend to propose answers to all or even most of the basic political questions that we ask. It is therefore compatible with lots of different ideologies, like nationalism, which it must abandon when pushed too far. The same is true with concerns about poverty. There are good reasons for liberals to pursue these concerns, some distinctly liberal reasons, some not. But there also good reasons for liberals to be wary of making the elimination of poverty the primary source of a regime’s legitimacy, as has been generally recognized in the wake of the Soviet experience. Even Rawls, after all, subordinates the difference principle and its concerns about economic insecurity to his principle supporting equal liberties. So I don’t think that fear of poverty fits very well into the framework for the history of liberalism that Kahan has proposed. And that would be clearer if he had devoted a little more space to broadening the genus of liberal fear, to lumping as well as splitting.