The Stoic Criterion of Knowledge

a picture of a pomegranate hanging on a tree

Epistle 22. Posted on 2019-06-24.

A graspable impression (Greek: phantasia kataleptike) is a sensory impression that contains information which may be regarded as knowledge, given proper reasoning. As such, grasping or katalepsis is the Stoic criterion of knowledge. This article presents my interpretation of the method of grasping a graspable impression, of discriminating knowledge from opinion in Stoicism.

Zeno of Citium (c.334 BCE – c.262 BCE), the founder of Stoicism, introduced katalepsis. In De Natura Deorum Academica (2.145), Cicero (106 BCE – 43 BCE) reported that Zeno illustrated katalepsis with a gesture:

he would display his hand in front of one with the fingers stretched out and say A visual appearance is like this; next he closed his fingers a little and said, An act of assent is like this; then he pressed his fingers closely together and made a fist, and said that that was grasping (and from this illustration he gave to that process the actual name of katalepsis, which it had not had before); but then he used to apply his left hand to his right fist and squeeze it tightly and forcibly, and then say that such was knowledge, which was within the power of nobody save the wise man.

In Against the Mathematicians (7.248), Sextus Empiricus (c.160 – c.210) — a critic of Stoicism — reported the criteria of katalepsis: a graspable impression is one caused by an existing object and imaged and stamped in the subject in accordance with that existing object, of such a kind as could not be derived from a non-existent object. The first two clauses are the original criteria of Zeno, and the third clause was added later by the Stoics due to arguing with the Skeptics (Empiricus, Against the Mathematicians, 7.253). Epictetus (c.55 – 135) also suggested testing impressions — with respect to the second clause, I infer. I suggest that the later Stoics who added the third clause did not understand Zeno’s katalepsis in the same way that Antipater of Tarsus — the sixth scholarch of the Stoa — did not understand Stoic logic when he proposed that one-premise arguments are reasonable.

I interpret the first two criteria of Zeno as requiring knowledge to come only from a sensory impression, and that the common and peculiar qualities of the definition of its concept must be evident within that same impression. Allow me to explain.

Of impressions (alternately called presentations, representations, or appearances), some are sensory impressions (Greek: phantasiai) while others are non-sensory impressions. The subject of any sensory impression (Greek: phantasia) was caused by the object of sensation (Greek: phantaston), and so a sensory impression is caused by an existing object. For example, if Dion sees a mirage and his unconscious brain misperceives that water appears on the horizon, the appearance of the water is still caused by an existing object, a body, albeit not by water. In contrast, a non-sensory impression (Greek: phantastikon) is not caused by an existing object, but by other impressions. Examples of non-sensory impressions include dreamt objects, imagined objects, thoughts, and so on. The causing object of a non-sensory impression is a phantasma, which is referred to as a figment, or a figment of imagination. Non-sensory impressions are also referred to as imaginary impressions and colloquially as false impressions. Every non-sensory impression is false because the appearance is not caused by the existing object that it represents. Knowledge, as you see, must come from a sensory impression.

The second criterion requires that the common and peculiar qualities of the subject of the impression must be evident within the same impression. For example, if Dion encounters a human and receives an impression of a human, Dion then defines a human as a rational animal, where animality is the common quality and rationality is the peculiar quality. These qualities must be evident within Dion’s impression of the human in order to regard it as knowledge. Luckily, the sensed human is talking, and this is evidence that supports the definition of rationality, of being capable of reason; the object cannot be talking understandably and be only irrational, such as an irrational animal. Therefore, the talking human is evidence of rationality, which is also evidence of animality, and finally is evidence of humanity. Dion grasps as knowledge the humanity of the subject in the impression.

The third criterion, which I hold as unimportant, has also been translated as of such a kind that it could not become false, or of such a kind that it could not arise from what is not. The kind is probably better regarded as a predicated category of a body, such as resulting from a definition. In short, this criterion requires that it is conceptually impossible that the definition does not apply to the subject. For brevity, I am also ignoring the comment by Chrysippus about preconception with respect to the critierion of truth, and preconception in general, again, only for the sake of brevity, even though preconception is an important consideration.

As a quick note, I understand that grasping is the best translation of katalepsis, and so some modern Stoics refer to a phantasia kataleptike as a graspable impression. A less popular alternative is comprehension, which also means grasping but occurs only in a mental context. A comprehensible impression is an acceptable alterative. However, the most popular translation may be a cognitive impression, because cognitive derives from I know. And there are many others, including the adequate representation (of Pierre Hadot), apprehensible impressions, and convincing sense-impressions, to name a few.

The most famous example of katalepsis is a tale regarding Sphaerus (c.285 BCE – c.210 BCE), one of the pupils of Zeno:

After making considerable progress in his studies, [Sphaerus] went to Alexandria to the court of King Ptolemy Philopator. One day when a discussion had arisen on the question as to whether the wise man could stoop to hold opinion, and Sphaerus had maintained that this was impossible, the king, wishing to refute him, ordered some waxen pomegranates to be put on the table. You have given your assent to [an impression] which is false. But Sphaerus was ready with a neat answer. I assented not to the proposition that they are pomegranates, but to another, that there are good grounds for thinking them to be pomegranates. Certainty of [impression] and reasonable probability are two totally different things.

In this example, Sphaerus may have thought the pomegranates looked suspicious, or may have been unsure of his definition of pomegranates which was most likely developed on the spot. Perhaps he knew that the appearance of a pomegranate was imitated easily with wax. Perhaps Sphaerus would have assented to a pomegranate being a pomegranate had he seen it hanging on a tree. For whatever reason, Sphaerus did not regard the impression as graspable. He assented to a different proposition, a subsequent thought, which is a non-sensory impression. The subsequent thought may be assented to and regarded as true and conceptually necessary, but it cannot be grasped as knowledge. This is noteworthy, because the information in a graspable impression is also regarded as conceptually necessary, due to definitions. To return to the example of seeing a human, consider the definition of humanity as a Chrysippan conditional:

If Both x is an animal and x is rational, x is human.

It is conceptually necessary that any body that is a rational animal is human, because it is conceptually impossible that Not such a body is human; An inhuman rational animal causes a (mental) conflict (Greek: mache). Therefore, the subject in the sensory impression must be human as perceived and with the strength of conceptual necessity. This judgment is strong and unchangeable, as long as the concepts and therefore premises of the argument do not change — as in a changing proposition (Greek: metapiptonta) or a changing argument (Greek: metapiptontes logoi). In short, it is implicit that all knowledge is conceptually necessary.

The contrary of knowledge (Greek: episteme) is opinion (Greek: doxa), which is weak and changeable. According to Sextus Empiricus, an opinion is a weak and false assent (Against the Mathematicians, 7.151). Anything that does not arise from sensation is merely an opinion. Any sensory impression that lacks the self-contained conceptual necessity of the subject due to definitions can yield only opinion, not knowledge. I interpret weakness as lacking conceptual necessity, such as being conceptually non-necessary and/or conceptually possible only, and that a false assent is assent to a proposition that lacks conceptual necessity such as when an indifferent has only circumstantial value rather than continuous, intrinsic value. Epictetus still recommends testing the qualities of the definitions of these impressions, and such testing requires other impressions; these are not self-contained. More specifically, Epictetus recommends testing sensory impressions (external impressions) that may perturb your peace of mind:

Make it, therefore, your study at the very outset to say to every harsh external impression, You are an external impression and not at all what you appear to be. After that examine it and test it by these rules which you have, the first and most important of which is this: Whether the impression has to do with the things which are under our control, or with those which are not under our control; and, if it has to do with some one of the things not under our control, have ready to hand the answer, It is nothing to me.

Or, to test every impression whenever possible:

Therefore, the first and greatest task of the philosopher is to test the impressions and discriminate between them, and to apply none that has not been tested.

Marcus Aurelius (121 – 180) was no different:

Persistently and, if possible, in every case test your impressions by the rules of physics, ethics, logic.

This epistemology cannot guarantee the absolute truth of knowledge; nothing can, because absolute truth is unknowable. It is always possible that a dreamt object is confused with an existing object, or that a definition was problematic. The strength of the Stoic epistemology is that it limits knowledge to apply only to the causing object of a sensory impression, and only when it is conceptually necessary that the causing object matches the definition of the perceived concept. I regard it as stronger than modern science, which uses modus tollens to falsify theories, but that also lacks any verification principle — which the logical positivists tried so hard to establish. I regard katalepsis as the Stoic verification principle, except that it separates knowledge from opinion rather than cognitively meaningful statements from cognitively meaningless statements.

Stoic knowledge is much more rigorous than merely requiring conceptual necessity. This application of conceptual necessity also requires a definition to be met, and it requires that the subject is also the causing object of a sensory impression. Therefore, Stoic knowledge applies only to physical bodies that have been sensed. Consequently, every non-sensory impression — including thoughts — can yield only opinion, not knowledge. Stoic knowledge is, then, a blend of empiricism and rationalism. Knowledge must be grounded in sensations of a body in the cosmos, and hence it is empirical. However, knowledge also requires conceptual necessity with respect to definitons of concepts, and so it involves rationalism.

Katalepsis is a skill that requires practice. In the beginning, a novice Stoic fumbles mentally to develop Stoic definitions. With practice, less time is required to develop a definition, and the accuracy of the definitions improves. In time, many causing objects may be defined accurately in a few seconds, though of course some impressions may confuse the novice Stoic for a few minutes, or longer. All of this is very acceptable. As any Stoic advances in Stoicism, it becomes easier to define the important concepts in life, as well as the concepts that one struggles with, with respect to peace of mind. Of course it is impossible to analyze every impression, but it quickly becomes very practical to analyze every potentially problematic impression that may perturb peace of mind.

There is quite a bit more to the topic of the graspable impression, but this article should be a sufficient introduction for any novice Stoic who seeks to embrace the Stoicism of the Old Stoa, which is to say, orthodox Stoicism. Developing definitions and analyzing impressions of everything important in one’s life is the very basis of the practice of Stoicism. The skill of katalepsis is useful for preventing unreasonable evaluative judgments because the causing object of a graspable impression cannot be regarded as a good or an evil; for an example, see Epictetus (Discourses, 3.8.1-4). Another important use occurs during introspection when one is attempting to correct one’s faults, to improve one’s character by contemplating an impression that participates in concepts when it should not, and to prevent assent in the future to the perception of those concepts applying to similar impressions; the skill of katalepsis is necessary for extirpation.

The Skeptics did not seem to understand the criterion of the graspable impression, and often attacked it. The graspable impression does not ensure absolute truth. Now that the reader understands the criterion, every Skeptic example may be met and overcome.

Skeptics do not deny appearances, but deny the possibility of knowledge referring beyond the appearance of an object to the object in the cosmos. It is always disputable if the object is as it appears to be in an impression of it. The best we can do is to consider the appearance in relation to other appearances, the present impression in question in relation to past impressions. Any object in nature can only ever be described as a collection of appearances, as a collection of properties, which are resemblances to other appearances. Stoicism provides a criterion about when to regard a perception of an appearance — an impression — as knowledge, but do not regard knowledge as equivalent to absolute truth. Instead, Stoicism regards knowledge as certainty given one’s well-reasoned experience, as a criterion of conceptual truth or experiential truth, rather than absolute truth. Even Chrysippus’ goal of life is tempered this way: To live in accordance with one’s experience of the things that come about by nature.

Go forth and scrutinize your impressions, so that you may live according to nature.

Vale (pronounced WAH-lay is Latin for Farewell),

Ron Hall

Ron Hall

See Also


Instagram Pinterest YouTube Channel

Call now to set up an appointment

(860) 960-6711*

*Mon-Fri,9am-5pm ET