Historical illustrations 49 Discernibility thesis. If one is attentive enough, one believes that one

has a cognitive impression if and only if one has.

Ur-foundationalism. One has basic cognition thatp iffone assents to a cognitive impression thatp.

One assents to a cognitive impression thatpiffone has a cognitive impres-sion thatpand one believes thatpon the basis of that impression. If the Discernibility thesis holds, Stoics took having a cognitive impressionto be discernible. Assuming basing facts and contents to be discernible, being based on a cognitive impression that pis a discernible mark of truth of one’s belief that p. By the infallibility thesis, it is also an infallible one. Stoic Ur-foundationalism is the idea thatbeing based on a cognitive impressionis a discernible and infallible mark of truth.

Cognitive impressions were standardly defined as follows:

A cognitive impression is one which arises from what is and is stamped and impressed exactly in accordance with what is, of such a kind that could not arise from what is not. (LS 40E, SVF 2.65; see also LS 40C, SVF 2.53; LS 40D)

Since cognitive impressions are “impressed exactly in accordance with what is”, they are true (Frede,1987, 164). So the infallibility thesis holds.

Cognitive impressions are also said to be “of a such a kind that could not arise from what is not”. As other texts make clear, Stoics meant by this that cognitive impressions were discernible from non-cognitive ones:

’Of such a kind as could not arise from what is not’ was added by the Stoics, since the Academics did not share their view of the impossibility of finding a totally indiscernible [but false]

impression. For the Stoics say that one who has the cognitive impression fastens on the objective difference of things in a craftsman-like way, since this kind of impression has a pecu-liarity which differentiates it from other impressions, just as horned snakes are different from others. (LS 40E, SVF 2.65)

It is not entirely clear which discernibility thesis the Stoics held. The two extreme readings are as follows. (1) There is a unique character C

that all cognitive impressions share, and a sufficiently attentive subject believes that an impression has C iff it has. For instance, all and only cognitive impressions are“clear and distinct” (LS 40C). (2) Each cognitive impression has some characterCi that no non-cognitive impression has, and a sufficiently attentive subject believes that an impression hasCiiffit has. For instance, when one looks at Socrates well enough and in proper conditions, one gets an impression of him with a distinctive character that no impression of something else has, and a sufficiently attentive subject believes that an impression has that character iff it has. Various intermediate views are possible. I shall not try to adjudicate, but I will assume the first option for simplicity and concreteness.³⁸

Assuming that cognitive impressions share a common character, we have some indications of what it was supposed to be. Cognitive impres-sions were supposed to be “clear and distinct” (LS 40C, SVF 2.53). What Stoics meant by “clear and distinct” is not made explicit, however.³⁹ They

38. Call the two readingsunitaristandparticularist, respectively. In support of the par-ticularist reading, we can point out the fact that Stoics argued for the discernibility thesis on the basis of their metaphysical view that non-identicals were discernible (“no hair or grain of said is in all respects of the same character as another hair or grain”, LS 40J).

They seem to conclude from the discernibility of non-identicals that any adequate (that is, detailed enough) impressions of two distinct objects would themselves be distinct. In support of the unitarist reading, some passages characterise the cognitive impressions in general as being “clear and distinct” (LS 40C, SVF 2.53) or “self-evident and striking”

(LS 40K). BothLong and Sedley(1987, 250) andFrede(1987, 162) adopt the unitarist reading, but without explicitly considering alternatives. Frede(1987, 162) writes: “the Stoics also seem to assume that cognitive impressions by themselves differ from all other impressions, that there is some internal characteristic that serves to mark them offfrom other kinds of impressions and allows the mind to discriminate between cognitive and non-cognitive impressions [...].” He interprets the “horned snake” metaphor above as an analogy for “some internal differentiating mark”, by which he appears to mean a unique mark shared by all cognitive impressions. Long and Sedley(1987, 250) adopt a similar position, and assume that clarity and distinctness is the unique mark, on the basis of LS 40C, SVF 2.53.

39. The fragment suggests that clarity has to do with the impression “arising from what is” and that distinctness has to do with the impression “being exactly in accordance with what is”:

“Of the impressions, one kind is cognitive, the other incognitive. The cognitive, which they [the Stoics] say is the criterion of things, is that which arises from what is and is stamped and impressed exactly in accordance with what is. The incognitive is either that which does not arise from what is, or from that which is but not exactly in accordance with what is: one

2.3. Historical illustrations 51 were also supposed to be “self-evident and striking” (LS 40K). Saying that

an impression is “evident” may mean thatits content(or the relevant as-pect of its content)is obvious: an impression of an ox is “evident” only if it is evident that it is an impression of an ox (Frede,1987, 159-160). Alterna-tively, “self-evidence” and the “striking” aspect refer to the same thing, namely that cognitive impressionsforce assent and action(LS 40H; LS 40K;

LS 40O). If their strikingness is to be distinctive, then it should be some-how special, for Stoics also allowed that some non-cognitive impressions were “convincing” (LS 39G, SVF 2.65).

Even though the details are blurry, the intuitive idea behind the Stoic discernibility thesis is clear enough: when one’s eyes are open, one is in a normal state, close enough and in a well-lit area, one can get an impression of an object of a kind that one could not get otherwise. And they thought that impressions of this kind could not be mistaken. The following passage argues that common sense agrees with Stoics on this point:

“So too, whenever someone is keen to grasp something pre-cisely, he is seen to chase after such an impression of his own accord, as when, in the case of visible things, he gets a dim impression of an object. He strains his sight and goes close to the visible object so as not to go wrong at all; he rubs his eyes

which is not clear or distinct.” (LS 40C, SVF 2.53)

One idea would be that clarity is a sort of vividness that results from an impression having been caused normally by an external object (as opposed to dreams, for instance), and that distinctness results from the impression being sufficiently detailed (as when one looks at an object from a close distance). But that is sheer speculation.

By contrast, Frede(1987, 159) glosses “clear” as follows (without textual support either): “under normal conditions we not only have an impression which does not misrepresent things but we have one which represents them clearly, that is, affords us a clear answer as to what kinds of objects we are facing. And under normal conditions we do in fact have a clear view of an object we are confronted with, and we can tell without difficulty what its visual features are. Let us call such an impression “clear”

and “evident”.”

Frede’s gloss on “distinct” (which draws on Stoic metaphysics and Stoic-influenced medical doctrines) fits the “sufficiently-detailed” reading: “a cognitive impression of an object will involve a representation of this object which is so articulate that the only object which will fit this representation is the very object the impression has its origin in” (Frede,1987, 162).

and does just everything until he takes in a clear and striking impression of what he is judging, as though he thought the reliability of the cognition rested on this.” (LS 40K)

Thus Stoics thought that when the conditions are appropriate, one can get an impression of Socrates of a kind that one could not get from a twin. They thought that when one is wide awake, one gets impressions such that one could not get while dreaming. More generally, they were committed to the idea that one could always discern whether one had an impression of the cognitive kind. This does not mean that one can always tell whether animpressionis true or false; but one can always tell whether an impression isof the cognitive kind or not. And when it is, it is guaranteed to be true. So, surprising as it may be to us, Stoics were committed to the claim that there are some impressions one gets when one perceives that cannot be like the ones one gets in dreams, and that there are some impressions one gets from seeing a certain man that could not be had by seeing his twin (see LS 40I, LS 40C andFrede,1987, 162–3).

Thus, Stoics endorsed the discernibility thesis.

Not only did the Stoics endorse the discernibility thesis, but it was an explicit premise in their debate with Academic sceptics thatif there were no discernible class of true impressions, we could not have knowledge. Here is Cicero’s reconstruction:

Zeno defined [a cognitive impression] as an impression stamped and reproduced from something which is, exactly as it is.

Arcesilaus next asked whether this was still valid if a true impression was just like a false one. At this point Zeno was sharp enough to see thatif an impression from what is were such that an impression from what is not could be just like it, there was no cognitive impression. Arcesilaus agreed that it was right to add this to the definition, since neither a false impression nor a true one would be cognitive if the latter were just such as even a false one could be. But he applied all his force to this point of the argument, in order to show that no impression

2.3. Historical illustrations 53