Shortcomings with considering firsthand knowledge as supreme

Question

I think it is really common for people to believe something just because they are told it, although there may be implicit contextual factors like if they consider the source of information worthy of acceptance or of suspicion and rejection.

My bland, uninspired, default theory for why this is is evolutionary.

If we are to believe that much of the behavior of the world as we understand it can reduce to logical principles, we can conjecture that there is a valid logical way to make inferences about the world (for example, unknown knowledge, or future states of affairs). This is similar to Laplace’s daemon: if someone was omniscient (had access to all the contingent facts of a world), and they knew the complete laws of logic, they have an effective method to answer any question (let’s set aside that there may be inherent limitations to the logic itself, such as the laws of physics being inherently probabilistic, or there being inherently undecidable propositions in a logical theory.)

So, while it is these logical laws that explain everything that happens in the world, and while an entity who knew those rules could make unlimited valid inferences about the world, humans do not tend to think in that way, but they do still tend to make inferences about the world. In line with the idea of cognitive heuristics, from Daniel Kahneman, we can envision that the laws of human thought are logically definable algorithms, yet which are logically unsound: they are not guaranteed to lead to correct inferences.

It is very common to hear that a lot of features of biological structures are simply due to the advantageousness of conserving energy; and/or that energy is a constraint on a biological process.

So, premise 1 is that the average examples of human thinking are not logically valid. People throughout history have analyzed recurrent “logical fallacies” humans commit, but we still don’t have a complete theory of correct reasoning.

---

From here, my argument is looser and in its infancy.

In trying to learn from my mistakes of erroneous judgment, I have often tried to classify categories of knowledge. I have often leaned towards the stance that first-hand empirical experience, barring reasonably identifiable extremal cases like dreams, hallucinations and delusions (which is not the topic of this question), is somehow the most fundamentally certain category of knowledge. To an extent, we do not even need to use predicative categories like “is true” or “exists” when describing immediate perception - we can merely say the contents of moment to moment experience are present. In a way, to claim that something perceived is “not true”, or “does not exist”, would be futile, because it would merely displace the meaning of “to exist”, and we would need to find a different word for whatever immediate experience “is doing” (if not “existing”). Carnap (I think) said that an analytic truth is one whose negation would be a contradiction. Thus, phenomenological experience is ineluctable. I do not know if this is what Descartes meant when he said cogito ergo sum.

By comparison, the fruits of reasoning or thought will always strike me as less undeniable, perhaps for many reasons. It is in some ways possible to simply “not think”, for a period of time - thus, thought itself is an element of moment to moment experience that can disappear, while moment to moment experience remains. It is also possible to think false, contradictory, or even incoherent thoughts. And lastly, it seems that to even attempt to justify certain kinds of thought, requires us to think about why some kinds of thought are valid. Therefore, we are forced to depend on and trust “thought” before we have even justified it; and it may be the case that we cannot find any justification that is not circular.

---

With the above as context, my question is about to what extent it is logically naive to think that you can doubt anything that is not experienced first-hand.

I mean this in a very applied context.

For example, I might say: “I am absolutely confident that I ‘have arms’, in the sense that I am feeling them (and in this case, it does not matter if we “live in the Matrix”, as it would be the same to say ‘this constructed entity in this Matrix perceptually experiences arms’). But I can never be certain that the Holocaust happened.” By denying firsthand experience of the events of the Holocaust, it can open hypotheses in my mind about in what ways the Holocaust could have been faked. I may ultimately only ever be able to have an inclination towards what I find likely.

I am interested in pretty strong logical refutations to the previous idea - an argument that there is a very compelling reason to believe in the capability to make inferences, that puts thought on equal footing with perception. Or, I am looking for reference materials to expand and embellish these thoughts.

The key idea is if it would be absurd to deny virtually everything that is not directly experienced, and why.

---

One paradigm I’ve been considering is that one cannot justify empirical knowledge through some of the standard premises of logic. Instead, “empirical knowledge” requires its own set of premises. In a way, the best hope one has for justifying empirical knowledge is empirically. So, we would look for an empirically justified theory justifying “belief in” empirical knowledge.

Answer 1

You assume 'a complete theory of correct reasoning' is simply out there to be found. But correct reasoning is more like building a bridge, than like the physics that governs whether the bridge holds.

'First-hand empirical experience' already usually involves consilience, the convergence of evidence, because we have multiple senses. Optical illusions rarely also involve noise say. Individuals have believed all kinds of crazy things, there have even been events of mass hallucination or psychosis. An insidious risk is post-hoc reasoning, where we come to a conclusion or decide on an answer, then work backwards to fit evidence.

So we like lots of people's experience to agree. But we also like to prefer a community of people with good standards of evidence, who use strategies and systems to minimise cognitive biases like post-hoc reasoning. Sound familiar?

I would argue that determining good evidence is what science is, and first hand experience is just the best opportunity for experiment. And that really examining how we get at good evidence is about intersubjectivity, pursuading someone else with what would pursuade us, leavened by scepticism and mutual engagement in trying to find truth.

Try philosopher Donald Hoffman on why we can't trust evolution to show us the world impartially: Do we see reality as it is? (Youtube link to TED Talk)

And neuroscientist Anil Seth on how many cognitive processes have to be stitched together outside of our awareness, before we can make any sense of the world, often editing memories and recollection after processing: Your brain hallucinates your conscious reality (Youtube link to TED Talk)

The damning criticism of Descartes' cogito is, you don't get to have those words, those abstractions, except by having been in community with other humans, who developed them over time together. A child raised by wolves with no human language simply could not experience or be told the realisation. All our knowledge of how the world works relies on trusting other humans. We could all bw wrong in substantial ways that matter, like say about the nature of time (it being imposed by our cognitive needs rather than being as we intuit it). But the solution isn't to give up on knowing anything you haven't experienced, even your own experiences need community to be understood, like language and education, and community done right is the best tool we have for establishing what has hapened and what is the case. The tools of science have emerged exactly to this, as has good historical scholarship.