The Three Kinds of Knowledge
Spinoza identifies three kinds of knowledge corresponding to three degrees of cognitive adequacy. The first, imaginatio, is knowledge through signs, images, and hearsay — passive reception of symbols without comprehension of their causes. The second, ratio, is knowledge through common notions — the intellect's grasp of structural regularities that hold universally. The third, scientia intuitiva, is intuitive knowledge — the direct perception of particular things in their necessary connection to the infinite substance, requiring biographical depth, embodiment, and mortality. The framework maps onto AI with unsettling exactness: large language models operate primarily through the first kind (pattern reproduction without causal grasp), exhibit significant capability in the second (identifying structural regularities across domains), and possess no demonstrated capability in the third. The implications are practical and immediate for education, work, and the specific forms of cognition that remain the human being's alone.
In the AI Story
The first kind of knowledge is what Spinoza describes as 'from having heard or read certain words we call things to mind.' It is the knowledge of the person who has been told the earth orbits the sun without understanding gravitational mechanics. The proposition is true. It is held inadequately. The person cannot extend it, modify it, or recognize when new evidence should update it. She has a fact. She does not have a truth. This is the knowledge-mode of the LLM by architectural necessity: the system processes patterns of symbols without comprehending the causal structures that produced them.
The second kind of knowledge — ratio — is where the machine's capability becomes genuinely valuable. When Claude synthesizes information across domains, identifying structural parallels between evolutionary biology and technology adoption, or between surgical innovation and software abstraction, it operates through what Spinoza called common notions: regularities that hold across contexts. This capability can accelerate the production of second-kind knowledge to a degree previously unimaginable. But the machine cannot reliably distinguish genuine common notions from spurious pattern-matches. The Deleuze error described in the adequate ideas entry was a spurious pattern-match that looked like a common notion. The human must test the regularities.
The third kind of knowledge is where analysis becomes most consequential. Scientia intuitiva requires something no current machine possesses: the experience of being a particular being with particular stakes in a particular world. The parent who senses her child is struggling before any behavioral evidence appears is not pattern-matching across a training set. Her perception is shaped by years of loving attention to this specific child. The senior engineer who feels that something is wrong with a codebase before she can articulate what — her perception is not abstract but the direct comprehension of this particular system, informed by thousands of hours of engagement. This is knowledge saturated with biographical specificity, and a being without biography cannot produce it.
The hierarchy is not merely taxonomic. Spinoza believed the third kind of knowledge was the source of the highest human joy — the amor intellectualis Dei, the intellectual love of God or Nature. The joy of intuitive knowledge is qualitatively different from the comfort of receiving smooth output from a machine. It is the specific satisfaction of having done the cognitive work that transforms a confused impression into a clear perception. This joy remains available in the age of AI, but only to those who resist the temptation to accept the machine's outputs as adequate ideas without doing the work of understanding.
Origin
The three kinds appear in Part II of the Ethics, Proposition 40, Scholium 2, and are developed through Part V. Spinoza's framework drew on Aristotelian distinctions between kinds of knowing while transforming them through his monist metaphysics. Medieval scholastic traditions had distinguished cognitio ab effectu from cognitio a causis; Spinoza radicalized the distinction by making the highest form of knowing not demonstrative syllogism but direct intuitive grasp.
The framework has seen renewed attention in contemporary philosophy of mind and epistemology, particularly in debates about tacit knowledge, expertise, and the limits of algorithmic cognition. Michael Polanyi's work on tacit knowledge, Hubert Dreyfus's critique of symbolic AI, and recent embodied cognition research all converge on distinctions that echo Spinoza's three kinds.
Key Ideas
Imaginatio as pattern without cause. The first kind is knowledge of symbols and their associations without comprehension of the causal structures that produce them — the architectural mode of the LLM.
Ratio as common notions. The second kind grasps structural regularities that hold universally; the machine's capability here can be genuinely extraordinary but requires human testing to distinguish genuine from spurious pattern-matches.
Scientia intuitiva requires biography. The third kind requires embodiment, mortality, and the accumulated specificity of lived experience — conditions a being without biography cannot satisfy.
Structural limitation, not temporary deficit. The machine's inability to produce the third kind of knowledge is not a problem more training data will solve; it is a consequence of the relationship between knowledge and the life of the knower.
The joy of understanding. Spinoza's amor intellectualis Dei — the intellectual love of Nature — is the specific joy available only through adequate understanding earned by the self, and cannot be outsourced without being extinguished.
Appears in the Orange Pill Cycle
Further reading
- Baruch Spinoza, Ethics, Part II, Proposition 40, Scholium 2.
- Baruch Spinoza, Treatise on the Improvement of the Understanding (c. 1662).
- Michael Polanyi, The Tacit Dimension (Doubleday, 1966).
- Hubert Dreyfus, What Computers Still Can't Do (MIT Press, 1992).