It must be thought that thought doesn't exist. I don't mean to engage in facile skepticism or to recycle a materialist propostion that thoughts aren't really real. Rather I mean to question whether the activity of thinking is autonomous, whether it has anything to do with something called thought, and if it does, whether thought can mean what we want it to mean.
If thinking exists it exists as a metapractice. Any practice can potentially be a metapracticereflective, abstract, critical, metaphoric, and yet no less a practice. No one field of practice is privileged in its metapractical function. Thinking is no more or less critical of cultivation than cooking is, for example. Secondary technical elaborations of practice can also take on a metapractical function, like paths between fields.
That thinking exists as an autonomous practice, however, may be iffy. What we call thinking may in fact be expression, calculation, composition, engineering, experimentation or the like, as the case may be. Perhaps no practice is completely autonomous, and we set the bar too high for thinking if we ask that it be autonomous. Does the idea of thinking as such require that we recognize an autonomy of thinking? The question of whether thinking is autonomous should not to be confused with the question of whether thinking must be about something. It's a matter of being able to recognize thinking for what it is, if that's possible.
We can generalize. Can generalization be decoupled from expression, calculation, composition, engineering, experimentation and the like? Do we believe our generalizations? This latter question should not be interpreted as assuming that thoughts aren't really real. It's a question that asks that we not make an assumption one way or the other. Perhaps our generalizations are intended to be hypothetical (or hypnothetical). If thinking is a passage, a bridge or a tunnel, it becomes a passage with an other side as a result of a decision. Stand on a bridge without deciding on a direction to take. Sooner or later the thought might arise that standing there thinking is a waste of a perfectly good bridge. Sooner or later we may decide to believe in our generalizations in order for the passage of thought to be a passage. In so doing, paradoxically, thought becomes something other than what it is, becomes a means to some other end.
Thinking may be a practice, and thought may exist. Thought engaged in a course of action may be no more or less real than thought apart from a decided upon course of action. Such indecisive thoughts, however, may not be the "products" of an activity of thinking, and not really a structure of thinking either. They fleet. We must consider whether there is a difference between thought as a metapractical field and thought such as it is. Since each may call for the other, we must also consider whether there's something like thought that bridges the two. I believe there's an aporia in there somewhere, but I will have to give it some thought.