[Topics: Assumption, Evidence, Skepticism, Belief]
The Least Assumptions:

Cartesian Skepticism, and Reducing Guesses and Assertions in a Belief Network to the Minimum


Portrait of René Descartes after Frans Hals - minimizing assumptions

Portrait of René Descartes (based on the painting by Frans Hals)


I closed the examination of pragmatic ethics in the previous article by saying that this time I would talk about “the one and only assumption I am always willing to make (and the only assumption that you should ever be willing to stand by).” So I’m going to do just that. But before getting to that one assumption, I want to make a few remarks about why it is important to minimize assumptions when forming beliefs.

As René Descartes famously observed, it is always striking how very much of what any given person claims to know (and so believe) rests upon a network of baseless or near-baseless assumptions, assertions, and heuristics so densely matted together that the person fails to realize that there is no actual solidity to its foundation whatsoever. An important feature of this nebulous nest of guesses and half-considered notions is the redundant and overlapping (if occasionally contradicting) nature of its constituent elements. It is just such a nest to which I aim to provide a superior alternative.

Assumption-Nests and René Descartes:

Just as the removal of a single twig or even a whole host of twigs from a bird’s nest will fail to compromise the structural integrity of the nest, so this mental nest can stand to sacrifice multitudes of assumptions in the face of legitimate criticism without dropping its insulated beliefs (i.e. its illusory knowledge). This sort of nest of assumptions is a comfortably versatile bed on which to rest beliefs which are convenient or agreeable, but which may not be defensible when proceeding only through reason and building with one fact at a time. Indeed, assumption-nests are often so messy and so tightly crammed with erroneous notions that they conceal from even the nests’ constructors the simple fact that there are no eggs within.

Human beings should be able to do considerably better than birds. Rather than just forcing together any random elements and hoping that they are smashed together in such a way that they will not be soon destroyed, humans can come to an understanding of the materials at their disposal; can analyze the conditions which their constructions will need to meet; and can build a proper structure. (Even if the goal is to incubate an egg and raise a baby bird, the human can improve on the design.) By comparison, the nest may be more malleable—and so resistant to piecemeal testing—but it is ultimately flimsy, in danger of being toppled by a light breeze or crushed by nature.

Crow Nest (Emőke Dénes) - René Descartes - minimizing assumptions

Photo by Emőke Dénes

Humanity’s physical outstripping of rudimentary construction should be matched by a mental outstripping of rudimentary belief. This is not to say, however, that unassailable beliefs can be found; any blueprint of a work of architecture is capable of being altered or improved in some way, and some seemingly sophisticated constructions can later be torn asunder by mistakes made in the design, or mistakes made when building, or unexpected scenarios arising.

The obvious question which follows all of this would be, ‘How would one find out which of their beliefs are well-founded, and which are faulty?’ This question can be answered in a number of ways. René Descartes provides one promising and immensely influential path. And another way follows: the first thing to notice is that a cursory confrontation of an existing belief might fail to reveal its insufficiency.

Confronting an existing belief lightly or conservatively would be like wanting to test the aforementioned bird’s nest without dropping (or, for that matter, seeing) the believed-in eggs. One begins pulling twigs out here and there and, after filling one’s hands and pockets with twigs without toppling the nest, one can pat oneself on the back and declare the belief’s foundation appropriately strong—generally followed by replacing the removed twigs, for future tests of a similar description.

The first concern of a human being who intends to claim that they know anything at all must be to think in a more radically simple way. If one imagines two versions of oneself simultaneously, one version as the thinker is now and one version stripped of every single scrap of knowledge except for rationality and sensory capabilities, what, if anything, could the former convince as true to the latter? Careful and simple methods of doubt and reconsideration like this one are some of Descartes’ greatest legacies.

The Least Assumptions and the Nature of Evidence:

I, like many thinkers before me, am much more impressed by René Descartes’ first two meditations than by the meditations that follow. The problem with leaving it at that, of course, is that agreement with only Descartes’ first two meditations, with no further thought, leaves a person mired in radical skepticism. Being thoroughly skeptical of any possibility of knowing about external reality is a prospective end to the conversation, as well as a prospective end to all thought of existence (in the existential, contextualized sense of the word).

So, for this reason—and following, as I have previously noted, from my adherence to a pragmatic and phenomenological philosophical orientation—I knowingly admit the entrance of one doubtful assumption into my network of beliefs and knowledge. That minimal assumption is as follows: although I will never be able to know beyond all possible doubt that this is definitely so, it is reasonable to act as if my senses are providing me with an at-least-semi-accurate representation of external reality.

However careful and couched, this is an assumption (in Kant’s phrase, a scandalous assumption), but I take it to be a necessary one. And, frankly, from a pragmatic and phenomenological point of view, the assumption is either true or else irrelevant to my perceptions; I have previously expressed this latter point by saying that I hold to no assumptions whatsoever. I think either expression paints a fairly accurate picture of my outlook.

Taking that as a foundational assumption, however, I endeavor (as Thomas Nagel counsels, in regard to the experience of absurdity) to retain my irony and lack of full commitment to that assumption. And I endeavor to make no others. I earnestly desire to rout my beliefs of all other baseless assumptions through the careful examination and evaluation of the available evidence. And I have respect for anyone who furthers the causes of reevaluation and self-examination with regards to the evidence for each belief held by themselves and others.

One person who has caused some stir and some indignation among internet logicians on the topics of overcoming biases and culling faulty beliefs is A.I. researcher and Harry Potter fanfiction author Eliezer Yudkowsky. For me, it is occasionally frustrating to read Yudkowsky’s articles; he oscillates between, on the one hand, being a careful thinker who restricts his assumptions and tries to appropriately match evidence to confidence and, on the other hand, fancying himself a strange guru figure and espousing a religious fervor for Bayesian rationality.

Thomas Bayes (allegedly) - René Descartes - minimizing assumptionsIn the end, though, there are some of his articles, like this personal favorite of mine on scientists who hold supernatural beliefs, which are lucid distillations of the hypocrisy involved in some forms of belief and some claims of knowledge—and the hypocrisy involved in restrictions on the scope of evidence as a concept. (And for good examples of the same writer at his least helpful, just follow either of the links in the ‘addendum’ at the bottom of that article.)

The essential kernel of that linked article can be found in this wonderful sentence near the middle of it, which perfectly expresses what I mean by ‘the scope of evidence:’ “Maybe our spiritual scientist says:  ‘But it’s not a matter for experiment.  The spirits spoke to me in my heart.’  Well, if we really suppose that spirits are speaking in any fashion whatsoever, that is a causal interaction and it counts as an observation.”

It’s this kind of work on evidence being broadly construed as any proposed justification that should be the focus of any conversation about the state of modern belief. This is a demand for clarity and rationality as much as it is a call for minimizing assumptions. So, much as I would prefer if he would temper his claims—or at least ally them with some consideration of culture, art, and epistemologies that are alien to his own—Yudkowsky’s perspective seems valuable to me.


In conclusion, I take this assumed starting point to be the phenomenological course of action; I take it to be the existential course of action; I take it to be the Nietzschean course of action (insofar as Nietzsche’s considerations of wills, desires, and instincts relate to later existential and phenomenological positions); and I take it to be the Cartesian course of action (though I would be hard-pressed to imagine a statement of this assumption in less elegant terms than the theological procession of Descartes’ later meditations).

Radical skepticism is pragmatically irrelevant to my perceptions, so much so that I am willing, now as ever, to proceed from having made that one assumption. Indeed, I privilege the epistemological positions that I have outlined here because this is the only assertion necessary to make them work; and, as I have clarified, in the realm of phenomenology even this one might be optional. Regardless, all epistemological competitors I have so far met require many more assumptions, in addition to this very same one. And whatever the chances of it falling to pieces later, I hope in the short-term to build a solid structure upon it—one that outlasts many nests.

Related Articles:

[Topics: Assumption, Evidence, Skepticism, Belief]
The Least Assumptions:

Cartesian Skepticism, and Reducing Guesses and Assertions in a Belief Network to the Minimum

was last modified: March 11th, 2016 by Daniel Podgorski
Bookmark the permalink.

React to this Article