176 Giacomo Romano
categorical distinction, although we naturally categorize artifacts differently from natural
entities. (Often people are convinced that certain natural items are artifacts and vice versa.)
Yet I have not argued for a different account. To propose a better alternative account
than the ones reviewed, I should avoid the fl aws discovered. I sketch my hypothesis of
for-ness and functional knowledge in the next section where I also clarify some ideas that
could be useful when giving a more satisfactory characterization of the Design Stance.
10.3 A Functional Stance for the Design Stance
The hypothesis that I put forward is mainly based on intuition, and is abstract and specula-
tive at best. It does not aim at replacing the Stance of Design; it aims rather to make it
fully coherent and complete. According to the Design Stance approach, we recognize
artifacts by means of an inferential process that draws on (unspecifi ed) assumptions about
the relationship between a certain object and the intentions that its putative designer has
for it.
I maintain that we do apply a cognitive scheme when we recognize that certain entities
are for something, but this scheme is different from the one assumed by the theorists whose
work I have scrutinized. Such a scheme makes us perceive the elements of two states of
affairs as bound by a catalyst. The catalyst realizes the connective link that relates the two
states of affairs and in effect is the item that implements the transformation of one state
of affairs to another. This functional scheme also determines a basic temporal sequence
according to which the elements that are involved in the states of affairs at stake change.
The sequence consists of three logically ordered stages: fi rst the antecedent state of affairs,
then the process of transformation (that is realized by the catalyst or vehicle), and fi nally
the consequent state. This sequence is different from a linear causal chain and cannot be
reduced to it. In a causal chain the basic units are discrete causal links occurring between
two events; the causal relata of an individual causal link are two.^13 Instead, in order to
perceive the for-ness sequence, we need to perceive all of the three stages as inseparable.
Grasping for-ness requires the identifi cation of three-phasic, irreducible basic units; there
are three relata of the pattern identifi ed by the functional scheme. Furthermore, while we
perceive of causation as occurring between events (or objects), we recognize for-ness as
a relational feature occurring between states of affairs and a specifi c discrete entity (the
catalyst).
For example, the claim “yeast is for leavening the dough” entails a ternary relation of
an antecedent state (when the dough has not yet risen), a consequent state (when it has
risen), and the process of transformation (through the yeast) from the antecedent to the
consequent. Clearly there can be an alternative description that sounds more plausible such
as “yeast causes the rising of the dough,” but this is a description of a direct connection
between two events. Instead, to claim that “yeast is for leavening” (although it is a strange