The replication argument for incompatibilism

Research output: Contribution to journalArticlepeer-review


In this paper, I articulate an argument for incompatibilism about moral responsibility and determinism. My argument comes in the form of an extended story, modeled loosely on Peter van Inwagen’s “rollback argument” scenario. I thus call it “the replication argument.” As I aim to bring out, though the argument is inspired by so-called “manipulation” and “original design” arguments, the argument is not a version of either such argument—and plausibly has advantages over both. The result, I believe, is a more convincing incompatibilist argument than those we have considered previously.

In recent times, it has become increasingly common for incompatibilists about moral responsibility and determinism to invoke what have come to be called manipulation scenarios—roughly, scenarios in which a given agent’s behaviour is somehow “set up” in advance by various powerful manipulators or designers working “behind the scenes”. The proponent of a “manipulation” argument for incompatibilism puts forward the given scenario, hoping to elicit the judgment that the relevant agent cannot be fairly blamed for doing what he or she does. He or she then contends, typically, that there is no relevant difference between the agent in this (non-responsibility) scenario, and any agent in a deterministic world; there is no relevant difference between one’s actions having been guaranteed in advance by powerful designers, or instead guaranteed by (otherwise similar) blind natural causes. Determinism thus precludes moral responsibility.

In my view, the basic insight behind the manipulation argument is important and compelling.1 Many compatibilists, however, have not been convinced. Once one appreciates precisely the way in which such “designed” or “manipulated” agents may meet their preferred conditions on moral responsibility, they maintain, it will no longer seem clear that such agents could not be responsible for what they do. The incompatibilist judges that no such agent could be responsible; the compatibilist disagrees. Such is, broadly speaking, the state of the debate concerning manipulation arguments.

In this paper, I wish to move the debate forward, not simply by defending the ordinary incompatibilistic judgment in a standard manipulation case, but by building on the manipulation scenarios considered thus far to provide a new kind of argument for incompatibilism. As we’ll see, though my argument is inspired by manipulation arguments, my argument is not a manipulation argument—or even an “original design” argument. In particular, my argument bypasses the most common objections to such arguments, while remaining just as powerful as those arguments—and arguably even more so.

In order to present my argument, I am going to tell a story—a story rather more elaborate than those usually told in these contexts. (The elaborateness of the story will help to address one of the most common complaints about manipulation arguments, viz., that it is unclear how we are to envisage the details of such cases.) The story is, I admit, fanciful—but it is no more (or barely more) fanciful than stories with which anyone working in these debates is already familiar. The story is, in a sense, a version of the “rollback” scenario depicted by Peter van Inwagen.2 In the rollback scenario, one is asked to imagine that God “rolls back” the state of the universe until just before an agent deliberates and makes a given decision. In this context, the assumption of determinism implies that, in every rollback, we get precisely the same result. The fundamental diagnostic element in the rollback scenario is therefore time: we continuously “reset” time to just prior to the agent’s decision—and thus we see something about the nature of determinism. In my story, however, I wish to spatialize the rollback scenario. That is, instead of considering 10,000 (temporal) replays of the same set of initial conditions, I wish instead to paint a scenario in which one discovers 10,000 (spatially separated) distinct actualizations of these conditions. In considering this scenario, I believe that we come across a particularly powerful argument for incompatibilism. For reasons that will become clear shortly, I propose to call it the replication argument.

First, I tell my story. I then draw out in more detail the incompatibilist argument it makes possible, and respond to some objections. Next, I contrast the argument I develop with the “manipulation” and “original design” arguments already present in the literature. I conclude by suggesting how the present argument plausibly constitutes an advance over such arguments.
Original languageEnglish
Pages (from-to)1341–1359
Early online date19 Aug 2018
Publication statusPublished - Dec 2019


  • incompatibilism
  • free will
  • rollback argument
  • compatibilism
  • moral responsibility
  • manipulation


Dive into the research topics of 'The replication argument for incompatibilism'. Together they form a unique fingerprint.

Cite this