The Myth of the Given

(Written after reading O'Shea's What is the Myth of the Given?.)

Here is my atempt to reconstruct the Sellarsian critique of the Myth of the Given (most famously laid out in the paper Empiricism and the Philosophy of Mind) as a I understand it.

The most basic version of the myth is the categorial one.

Stated like this, the concept of the Myth of the Given is founded on the concepts of direct awareness and categorial status. An item in this concept, is, I think, just some mental event/thing/quale. An item has a categorial status C if it is a member of a particular (proto-/ur-)conceptual category (e.g. red items, round items, loud items). Direct awareness is about having a direct access to those features and those features being, in a sense, irreducible to other features the most basic ones.

Sellarsian critique of the myth, then, contends that there is no such a thing as most basic features. There is no privileged, one true most basic ontology or conceptual schema. The human conceptual schema is malleable and adaptive. The process of human cognitive development is not about finding abstractions over the basic sensory data, but rather about shaping the mind as a whole.

Sellars also thinks that for any mental element to count as knowledge, it needs to be a citizen of the logical space of reasons. What, then, is the relationship between this space of reasons and whatever things are being delivered by the senses? I am inclined to suspect it might be analogous to the relationship between means/actions/intentions on the one hand and ends on the other in Neo-Kantian philosophy of action (e.g. Korsgaard, 2009). An action can be endorsed as a means towards some end but its endorsement is not reducible merely desiring that end. Similarly, a mental element can be constituted as knowing something grounded in external senses or justifying them, but that knowledge is not just reducible to sensory activity (perhaps there's even some sort of duality at play here).

One might say that even though the categories/(proto-/ur-)concepts such as (a certain) color or (a certain) shape are learned/developed and in that sense achievements of human cognition, rather than given and basic, their convergence and stickiness gives them a certain sort of privileged status in comparison with, e.g., more malleable concepts, such as biological categories¹ or countries. Thus, we might say that human minds develop their conceptual schemes to be constrained in a certain way (due to a limited neuroplasticity later in life (short developmental window), a stable environment, etc).

My shoulder Sellars replies that it doesn't matter because the point of the notion of givenness or awareness-directness he's criticizing is not stability of the conceptual scheme but its primordial basicness and closeness to noumenon that gives certain sensibilia a direct access into the space of reasons being instantiated within the mind.

Tentative analogy: The membrane of the mind has no passive transport channels. All transport is active.

There is actually (perhaps) another critique of the doctrine of givenness that Sellars seems to dance around but doesn't talk about very directly.

Cognitive processes are, in general, in various ways, normative.² The normativity imposed on an agent qua agent is that it continues to be an agent and that it continues to be the kind of agent it is (even though "what kind of agent it actually is" is itself subject to reinterpretation). The normativity imposed on an agent's action qua agent's action is that the action achieves the desired aim in the desired way and does not cause other effects that the agent diswants it to cause. Finally, the informativity imposed on perception qua perception (of any modality) is that it appropriately latches onto (registers) some aspects of the world.

Crucially here, any kind of normativity requires/presumes the possibility of failure. One can fail to act if one doesn't act as one wanted to act or if the act fails to achieve the desired result. The agent can fail its own agency if it falls apart as an agent or turns into something that is not itself and not endorsed by itself for any reasonable™ interpretation of "self". And, obviously, one can mis-perceive. One can mis-see something that is not there, even if one was fairly convinced that this thingy that one thought one saw actually was there.

The doctrine of givenness, however, postulates a ground of absolute certainty, thus procluding the possibility of failure on that ground, which removes its normativity. Therefore, sense-datum-sensation is no longer normatively constrained and no longer an instance of perception.

Several things to note:

• This critique also applies to any sort of doctrine that is supposed to categorically deliver unmistakeable knowledge, e.g. Descartes's project.
• This critique does not apply to situations where (estimated/subjective) certainty is achieved contingently, rather than categorically, e.g. one might be justified to believe a proposition with probability $1$,³ e.g. given a mathematical proof of some proposition.
• There is an open question of: How should one carve out the boundaries between the domains, so that it is more clear when an instance of certainty is categorical rather than contingent?