As you scroll through this review,
you moveyour hand; this causes the mouse to move; in turn this causes,
via a series of intermediary events, changes on your screen. A bit more
reflection shows that this case is entirely mundane: causal relations are a
ubiquitous feature of the physical world. Causal relations are also, according
to many philosophers, at the center of phenomena like knowledge, perception,
linguistic meaning, mental content, belief, free action, and right action. In
fact, one is hard put to think of an important philosophical notion that has
not received a causal analysis, especially in recent analytic philosophy.
Consider a few from the theory of knowledge. According to the causal theory of
knowledge, knowledge is true belief caused by what makes the belief true. Or,
according to a competing view, knowledge is true belief caused by a reliable
belief forming process, where a reliable process is one that causes a high
ratio of true to false beliefs. According to the causal theory of perception,
seeing that the cup is on the table consists in being in a perceptual state
that is appropriately caused by the cup and the table. Causation, it seems, is
absolutely central. We will need to understand causation itself if we are to
understand either causal theories in philosophy or the nature of the
surrounding world.

            Working
through the papers in Causation and Counterfactuals will help us with
this. The volume consists in eighteen cutting edge papers (twelve new, six
previously published) by the best people in the field, as well as an editors’
introduction. Most are devoted to one leading view of causation — the
counterfactual view. Hume articulated the basic idea this way: "we may
define a cause to be an object followed by another . . where, if the first
object had not been, the second never had existed." (An Enquiry
Concerning Human Understanding, Section VII) Intuitively, this is the view
that causes make a difference, enough of a difference that had they not
occurred their effects would not have occurred either. On this view, what makes
it true that the ingestion of the poison killed the spy –that is, what makes
it true that ingesting the poison caused the death of the spy — is the
following fact: ifthe spy had not ingested the poison, then he would
nothave died. This last statement is what philosophers call a
counterfactual conditional: it is conditional because of its distinctive
if-then form; it is counterfactual because, roughly, the part following ‘if’
(namely, ‘the spy didn’t ingest the poison’) is counter to fact (i.e. false).

            In applying
this view to see if it provides us with the right verdict about whether some
event, c, is the cause of some distinct event, e, we must have a consistent way
of determining exactly what the world would have been like had c not occurred.
It is only when we know exactly what the world would have been like minus c
that we can then go on to determine whether, as the theory alleges, e would
have been absent as well. To do this, David Lewis, the leading contemporary
proponent of the counterfactual theory, employed the idea of a divergence
miracle. If c occurred at t, then in considering a possible world in which c
does not occur, we are to consider a world in which the laws of the actual
world are violated shortly before t in a localized way that is just sufficient
enough to permit c’s not occurring. Then imagine the remaining events fall into
place and look for e: if e is not present, then c made enough of a difference
that without it, e would not have occurred. This qualifies c to be a cause of
e.

That is the basic idea. In filling
in the details of this or any other account of causation, one faces two
challenges. Both are formidable. The first consists in an ever-expanding body
of difficult puzzle cases (e.g. cases of redundant causation, prevention,
double prevention, early and late preemption, delaying and hastening, trumping,
causation by absences, various kind of probabilistic causation, and more). The
cases are so numerous that the editors decided to include an index to help
readers keep track of the 200-plus cases discussed in the various papers in
this collection. The other challenge is to deliver a set of illuminating
explanations about various features of causation. These divide into
controversial features and uncontroversial features. For the controversial
features, an illuminating explanation, but no particular verdict, is required.
Here we find the possibility and nature of backward and simultaneous causation;
the possibility and nature of causation by, and of, absences; whether causation
is transitive; and the role of causation in chancy processes. Other features
are uncontroversial. Here, both an illuminating explanation and a particular
verdict is required. This is where we find the asymmetry of the causal
relation, probabilistic causation, and the difference between causing an
outcome and guaranteeing it.

            The papers
by opponents of the counterfactual view fall into three groups. Two are focused
on raising problem cases. Here Jonathan Schaffer raises novel cases of trumping
preemption; and John Collins raises various prevention and trumping cases.
Another paper relies less on intuitive cases: Ned Hall argues that
counterfactual views fail to tell us in a satisfying way whether causation is a
transitive relation. A third set of papers is devoted to developing alternative
accounts and showing that they are, in some way, superior to the counterfactual
view. Here, Tim Maudlin argues for an approach that takes laws of nature as
central and primitive; D.M. Armstrong argues for a singularist view that takes
causation to be a primitive relation between universals; Ned Hall argues (in a
second paper) for thinking that we really have two quite different concepts of
causation (one captured by the counterfactual view, the other by something Hall
calls ‘production’); and Igal Kvart argues for the superiority of a
probabilistic view. In another paper, Christopher Hitchcock raises some novel
challenges to the last view.

Friends of the counterfactual view
are primarily concerned with developing the counterfactual view so that it can
handle various objections. Sometimes, as with the papers by Stephen Yablo and
Peter Menzies, the counterfactual view is taken in directions quite different
from the one Lewis took it in. Into this group fall two papers by Lewis (one,
"Void and Object," is previously unpublished), the papers by Yablo
and Menzies, as well as papers by David Coady and Murali Ramachandran.

Last, a number of papers are
primarily concerned with defending views about the apparent relational nature
of causation. Here, L.A. Paul defends the view that the relata are property
instances; Lewis defends the view that absences (i.e. the non-occurrence of
events) can serve as causes; Helen Beebee argues that Lewis is mistaken about
this; Cei Maslen argues that causation is really a three-place relation between
a cause, an effect, and a contrast; and D.H. Mellor argues that causation is
not a relation at all. 

            As this
rundown indicates, not all the papers are primarily concerned with the
counterfactual view. This makes the collection less unified than its title suggests.
Still, it embodies the best of analytic philosophy — sharply focused
well-conducted constructive debates that go far beyond casual suggestions.
Often the interplay of objections and available defenses is played out quite
exhaustively. I suspect that along many of these roads, non-experts will be
left behind after just a few basis moves and countermoves have been made. But
even these aborted trips will be instructive. Further labor, and lots of it,
will be needed to reach the level of sophistication displayed in many of these
papers.

©
2005 Peter Murphy

Peter Murphy, Ph.D., is a lecturer in the department
of philosophy at the University of Tennessee-Knoxville.

Categories: Philosophical