Epistemic Collapse

13 April 2017


Not long ago in Snowstorm Reflections on Collapse and Recovery I discussed some of the experiences likely to be related to a local and limited collapse of social institutions, as a way to consider broader and deeper scenarios of social collapse. In this connection I quoted the following from Joseph Tainter’s The Collapse of Complex Societies:

“Collapse, as viewed in the present work, is a political process. It may, and often does, have consequences in such areas as economics, art, and literature, but it is fundamentally a matter of the sociopolitical sphere. A society has collapsed when it displays a rapid, significant loss of an established level of sociopolitical complexity. The term ‘established level’ is important. To qualify as an instance of collapse a society must have been at, or developing toward, a level of complexity for more than one or two generations. The demise of the Carolingian Empire, thus, is not a case of collapse — merely an unsuccessful attempt at empire building. The collapse, in turn, must be rapid — taking no more than a few decades — and must entail a substantial loss of sociopolitical structure. Losses that are less severe, or take longer to occur, are to be considered cases of weakness and decline.”

Joseph A. Tainter, The Collapse of Complex Societies, Cambridge: Cambridge University Press, 1988, p. 4

For Tainter, collapse is sociopolitical collapse, but we need not be limited by this stipulation. There are potentially many different meanings of “collapse” and I would like to particularly focus on what I will call epistemic collapse, which has played at least as prominent a role as social collapse in the extinction of civilizations.

A definition of epistemic collapse, that is to say, a catastrophic loss of knowledge, can closely parallel Tainter’s definition of social collapse, like this:

A society has epistemically collapsed when it displays a rapid, significant loss of an established level of knowledge (epistemic complexity). The term ‘established level’ is important. To qualify as an instance of collapse a body of knowledge must have been at, or developing toward, a level of complexity for more than one or two generations. The epistemic collapse, in turn, must be rapid — taking no more than a few decades — and must entail a substantial loss of epistemic structure. Losses that are less severe, or take longer to occur, are to be considered cases of epistemic weakness and decline.”

Tainter emphasizes that a “collapse” implies a previous level of attainment and stability (continuity); I agree with Tainter that this is an important qualification to make. It should also be pointed out that collapse implies a subsequent stability of the lower level of complexity and attainment, perhaps for a generation or two. In other words, a collapse — whether social, epistemic, or otherwise — means that stability and continuity at a higher level of complexity and integration is rapidly replaced by stability and continuity at a lower level of complexity and integration.

We know that one of the reasons the European “Dark Ages” were dark was the loss of the accumulated knowledge of classical antiquity, or, if not the loss (in an absolute sense), its restricted access due to loss of educational institutions, reduction in the publication, copying, and distribution of books, reduction in literacy, and so forth. During this period of reduced access to knowledge, some knowledge was lost in an absolute sense. Some books deteriorated or were destroyed before they were copied, and so have been lost to history. Much of the tradition of educational institutions was lost, as the educational institutions of classical antiquity went extinct or were extirpated (Justinian ordered the closing of the philosophical schools of Athens in 529 AD) and were subsequently replaced by educational institutions attached to the Catholic Church.

To reach further back into the past, around 1200 BC there was a generalized collapse that led to the extinction of several Bronze Age civilizations (this story is recounted in Eric Cline’s book 1177 B.C.: The Year Civilization Collapsed). This severe blow to civilization led to a significant epistemic collapse characterized by widespread loss of literacy throughout the ancient world. Homer, we recall, was recounting an “ancient” time of heroes and heroic deeds, and it has been speculated that the Homeric corpus was the translation into written form of oral poetry that survived from this dark age of more warfare and less reading as compared to the age that preceded it.

In the kind of generalized collapse resulting in the extinction of civilizations that characterized the Late Bronze Age, there was both social and epistemic collapse, but to what extent are these two modalities of collapse separable? Even if not instantiated in human history, is it possible for a civilization to remain socially stable while experiencing epistemic collapse, or to remain epistemically stable while experiencing social collapse? I think that counterfactuals could be constructed to illustrate the possibility of isolated social or epistemic collapse, but these would not be very convincing without some historical parallel to make the point. A possible example could be the destruction of the Library of Alexandria, which was not tightly-coupled to a social collapse, but which entailed a significant epistemic loss, or the Mongol destruction of Baghdad in 1258, which, again, was not tightly-coupled to social collapse (except for the collapse of Baghdad itself) but was a disaster for learning and certainly issued in permanently lower levels of epistemic attainment in the region. For an illustration of the opposite isolation, it is arguable that Byzantium preserved the epistemic record of Roman civilization even as all Roman social institutions collapsed and were replaced.

The above considerations suggest that a distinction should be made between collapse (of some particular kind) and the extinction of a civilization. Only the most generalized collapse over several classes of human endeavor result in the extinction of civilization, and we can obtain a more finely-grained appreciation of how societies ultimately fail and civilizations go extinct (or resist extinction) by separating social, financial, legal, religious, and epistemic collapse, inter alia.

Multiple collapses result in the extinction of civilization. Civilization is itself a complex institution that is comprised of many sub-institutions; that is to say, civilization is an institution of institutions. We can classify the institutions that go on to make up a civilization as social institutions, economic institutions, legal institutions, epistemic institutions, and so on. All of these institutions are intertwined in civilization, but it sometimes happens that even an integrated institution within civilization will collapse without the civilization of which it is a part collapsing. The many intertwined institutions that together constitute civilization mutually support each other and can bring a civilization through a difficult time if enough of these institutions persist despite the failure of other institutions.

If our nascent scientific civilization were to experience an epistemic collapse, but the social institutions of our civilization retained a significant measure of continuity, our civilization could enter into a state of permanent stagnation (something I noted as the greatest existential risk of our time in Where Do We Come From? What Are We? Where Are We Going?). If, on the other hand, we provide a robust backup of our knowledge, so thorough that a social collapse is not also an epistemic bottleneck, we could see the social institutions we know disappear even while our knowledge was largely intact and propagated into the future. Thus the human future itself admits of possible isolated social or epistemic collapse. Something like our civilization would survive on the other side of this collapse, after the recovery or replacement of the failed institutions, but that civilization would be fundamentally altered by the process.

. . . . .


. . . . .

Grand Strategy Annex

. . . . .

project astrolabe logo smaller

. . . . .

Epistemic Hubris

21 November 2010


The peacock is not a bird to hide its light under a bushel.

In More Evidence for the Copernican Principle I finished with this observation:

Now we know, and can demonstrate, that planetary systems are not unique to the Milky Way. From this stronger inductive position, we can with greater confidence extrapolate our existing knowledge to the furthest reaches of the universe.

The Copernican Principle tutors us in metaphysical modesty, but the growing evidence for the Copernican Principle, and the paucity of counter-examples, inspires us to metaphysical ambition. Scientific knowledge is the expression of this metaphysical ambition as much or more than it is an expression of metaphysical modesty.

As soon as I wrote this I realized that this is an idea that deserves its own independent exposition, as there is much that can be said on this head. I will touch on some of these issues here, though a full treatment would require a treatise, so I may need to return to this fascinating topic at a later date in order to refine and extrapolate my formulations as presented below.

I linked the above quoted idea to my post on Metaphysical Modesty, in which I discussed Jeffrey L. Kasser’s lectures on the philosophy of science published by The Teaching Company, and his exposition of the role of humility in scientific knowledge. There I wrote, “The Professor characterizes metaphysical modesty as, ‘The way the world is does not depend on what we think about it’.” And I added, “Now, this is simply an alternative formulation of realism, but Kasser has chosen to express realism as a moral virtue, and particularly as the moral virtue of metaphysical modesty.”

In recognizing the role of humility in scientific knowledge, and formulating it in moral terms, Kasser was not putting himself out on a limb, but on the contrary was staking out a classic position in the philosophy of science. Despite the contempt for philosophical ethics found in much early twentieth century positivist thought (and the formulation of doctrines like the emotivist theory of ethics), many of these scientifically-minded philosophers gave expositions of scientific knowledge saturated in moral significance.

While Bertrand Russell was never a positivist per se, nor simpliciter, he provides a wonderful example, perhaps even the locus classicus, of moralized scientific epistemology. Russell writes of the notion of good and evil being “extruded” from scientific philosophy. After an extensive explanation of how ethical preoccupations have compromised philosophical and scientific inquiry (in the last paragraphs of section I of “On Scientific Method in Philosophy”), Russell begins section II as follows:

“If the notion of the universe and the notion of good and evil are extruded from scientific philosophy, it may be asked what specific problems remain for the philosopher as opposed to the man of science?”

“On Scientific Method in Philosophy,” section II, collected in Mysticism and Logic

Russell’s criticism of the moral preoccupations of earlier philosophers is in the same paper:

“The ethical element which has been prominent in many of the most famous systems of philosophy is, in my opinion, one of the most serious obstacles to the victory of scientific method in the investigation of philosophical questions.”

For Russell, ethics is regulative of scientific thought, rather than constitutive of scientific thought, but that moral concerns are still present is unquestionable, as we see in his discussions of scientific humility:

“A truly scientific philosophy will be more humble, more piecemeal, more arduous, offering less glitter of outward mirage to flatter fallacious hopes, but more indifferent to fate, and more capable of accepting the world without the tyrannous imposition of our human and temporary demands.”

the last sentence of his “Mysticism and Logic” paper

And again:

“The concept of ‘truth’ as something dependent upon facts largely outside human control has been one of the ways in which philosophy hitherto has inculcated the necessary element of humility. When this check upon pride is removed, a further step is taken on the road towards which a certain kind of madness—the intoxication of power which invaded philosophy with Fichte, and to which modern men, whether philosophers or not, are prone. I am persuaded that this intoxication is the greatest danger of our time, and that any philosophy, which, however unintentionally, contributes to it is increasing the danger of vast social disaster.”

Bertrand Russell, A History of Western Philosophy, Chapter XXX, “John Dewey,” p. 828

And again:

“By the practice of methodological doubt, if it is genuine and prolonged, a certain humility as to our knowledge is induced: we become glad to know anything in philosophy, however seemingly trivial. Philosophy has suffered from the lack of this kind of modesty.”

Bertrand Russell, Our Knowledge of the External World, third from the last paragraph of the last chapter.

Would it be too much to say that scientific humility was a preoccupation of Russell’s? As I noted above, Russell is the locus classicus here, and Kasser was on firm ground following his lead.

Russell makes a persuasive case for the role of humility in science, but as I realized as I was writing about the further evidence we now have for the Copernican Principle, the role of ambition in science is no less central, and perhaps more interesting. As our patient methods of induction increase our level of certainty about an hypothesis, we rightly become more comfortable with its further generalization and extrapolation.

To make a sweeping generalization about a law of nature, as when Newton posited universal gravitation, is an act of epistemic hubris. That the mind can capture, in an act of thought, a truth that is as true immediately beneath our feet as it is on the other side of the universe, is nothing short of astonishing. Nevertheless, it can be expressed with cool detachment, as with Newton’s law of gravitation:

“Every particle of matter in the universe attracts every other particle with a force that is directly proportional to the product of the masses of the particles and inversely proportional to the square of the distance between them.”

Newton here invokes “Every particle of matter in the universe” without qualification. It is this simplicity that gives general laws of nature such great power of prediction and theoretical unity, but it also must be recognized as a triumph of epistemic ambition, if not epistemic hubris. As it often commented in regard to Newton, he said that “I feign no hyptheses” (“Hypotheses non fingo”) even while formulating an unconditional and universal law of gravitation. If this isn’t an hypothesis, I don’t know what is.

To contemplate the possibility of metaphysical ambition co-equal with metaphysical modesty as one of the springs of science brings us to the locus classicus of ambition, MacBeth’s speech as he contemplates politically-motivated murder:

I haue no Spurre
To pricke the sides of my intent, but onely
Vaulting Ambition, which ore-leapes it selfe,
And falles on th’ other.

Shakespeare, William, 1564-1616. Macbeth (1623 First Folio Edition)

Shakespeare’s use of “Vaulting Ambition” in this speech has often been quoted. I cite it here because of what it implies: a leaping-over of all that intervenes into order to get directly to the object without recourse to the painstakingly slow method of creeping along the ground. This way of formulating Vaulting Ambition reveals it as a non-constructive strategy, and this makes it interesting.

If you have not studied philosophy, logic, or mathematics you are not likely to be consciously aware of the formal distinction between constructive and non-constructive methods, or, formulated metaphysically, between idealism and realism. Nevertheless, the distinction is fundamental, and even those who cannot distinguish a constructive proof from a non-constructive proof will be immediately familiar with the intuitive instantiations of these divergent attitudes, as in “seeing is believing” (a constructivist idea) or “there is more to the world than we can see” (a Platonic, and therefore a realist, non-constructive idea).

Vaulting ambition in science, as revealed in breathtaking leaps of deduction to striking and unexpected conclusions, is usually a non-constructive enterprise. Non-constructive proofs are fascinating, and show us things we would probably not otherwise even guess, but they have their weaknesses. Some non-constructive proofs prove things but do not show us how to find them, construct them, or otherwise submit them to immediate observation, inquiry, or further analysis. For example, we may “feel in our bones” that there is more to the world than meets the eye, but not be able to say exactly what it is that the world consists of but which cannot be seen.

By way of contrast, the humility in science, of the sort recommended by Bertrand Russell and Kasser (though formulated in the language of metaphysical realism), is usually a constructive enterprise, whereby we reach our conclusions by the most slow and painstaking methods, so that when we arrive at our conclusion we know exactly how we got there, what we found, and we can point to the result of our research so that others can inspect it for themselves.

Both humility and hubris are to be found in scientific thought, with now one, now the other, taking precedence in the way we understand the world, but even when one is in the ascendancy, the other is never absent.

. . . . .

. . . . .


. . . . .

%d bloggers like this: