Seven Views of Contemporary Compatibilism 183more extensively on his work on consciousness in Consciousness Explained
(1991) and the nature of evolution in Darwin’s Dangerous Ideas (1995).
In Elbow Room, Dennett argued for the legitimacy of folk psychological
notions in the explanation of intentional action. His view invokes a range of
stances adopted toward a system, including the most basic physical stance, the
intermediate design stance, and the mentalistic intentional stance. The adoption
of a stance is justified by its effectiveness in understanding, predicting, and inter-
acting with the system. Key to free will is the intentional stance. According to
Dennett, even a thermostat can be interpreted from this stance as a very limited
intentional system since its behavior can usefully be predicted by attributing to it
adequate beliefs and desires to display it as acting rationally within some limited
domain. For example, the thermostat desires that the room’s temperature (or the
engine’s internal temperature) not go above or below a certain range. If it
believes that it is out of the requisite range, the thermostat will respond appropri-
ately to achieve its desired results.
One might worry that a thermostat does not really have intentions, by contrast
with titmice, toddlers, or college sophomores. According to Dennett, this
concern starts one down the wrong path (1973: 155). To seek a clean distinction
between some metaphysically authentic intentional beings from simulacra like
thermostats presupposes that there is more to any intentional system than adopt-
ing a stance toward it as an intentional system. If that stance genuinely pays
off—if it facilitates a fruitful exchange, allows for helpful predictions, allows
one to engage rationally with it—then it wins the status of an intentional crea-
ture. No special metaphysical tag is needed. Hence, for Dennett, the propriety of
adopting the intentional stance toward a system is settled pragmatically in terms
of the utility of its application in interacting with the system. Along with this
thesis goes Dennett’s claim that folk psychological explanations (appealing to
the intentional stance) are entirely consistent with more basic design and phys-
ical stances, the former appealing to the intentions, not of the system, but of its
designer, the latter appealing only to the basic mechanistic processes that cause
the system from moment to moment to move from one physical state into
another. Once a system becomes sufficiently complex, as with even a chess
playing computer, the intentional stance will become indispensable for success-
ful interaction (1973: 154).
Just as the decision to adopt toward a system the intentional stance is a prag-
matic one, so too is it a pragmatic decision to adopt toward a system the stance
that it is a morally responsible person. Dennett calls this latter stance the per-
sonal stance (1973: 57–8). As with the intentional stance, there is nothing meta-
physically deep required to interpret legitimately a system as a person (no special
faculty of the will, for instance). Such systems are legitimately regarded as
morally responsible agents if interpreting them according to the personal stance
pays off (1984: 158–63). And just as the physical stance—which may be
deterministic—is compatible with the intentional stance, it is also compatible
with the personal stance. Furthermore, Dennett argues that it is practically
impossible to interpret and predict the kinds of complex systems under