s a provocation for this month’s meeting, I would like to pose two questions. I will explore these in more detail when we meet; but for now, I’ll give a few hints of some paths our discussion might follow.
The first question I’d like to explore is: do we take explanations seriously enough? This summer will be my fourth year as a participant in Re:Enlightenment meetings, and during each session that I’ve attended there has been significant disagreement over the nature of knowledge—what it is, how it works, and to what extent our explicit theories about it do, or should, influence our work. Cliff and I often claim that the fundamental value in the growth of knowledge involves the creation of good explanations, and it is this idea that I’d like to reflect on during our meeting.
The second question is: is Re:Enlightenment essentially a conservative project? Or perhaps I mean: should it be a conservative project? I’m not thinking of conservatism in a political sense, but rather in an epistemological sense, and, perhaps more broadly, in an intellectual sense. Part of the reason for my question is simply casual curiosity about how we think about that “Re:”, which simultaneously looks forward and backwards. And, of course, reference to the “Enlightenment” points backwards as well, especially considering how many of the members are historians, and how much the group seems to emphasize the “sitedness” of knowledge.
The two questions are, in fact, related. One of the points I would like to make is that taking explanations seriously means taking existing explanations seriously—and this often means sticking with explanations, even when we suspect that they are wrong, if we don’t have an alternative fundamental explanation that we prefer. I have a difficult time imagining a more conservative way to describe the workings of knowledge than this, which privileges existing theories until a better successor is available. (And of course, as our theories improve, they become stronger, more complex, and thus more difficult to replace…and this means that epistemological conservatism may itself become stronger as time goes by.)
This way of thinking about the importance of existing explanations is familiar in the sciences. Take general relativity and quantum mechanics. Both theories are at odds with each other, and as a result we know that at least one of them is false. (Of course, in the long run, both are false, as are the rest of our explanations…but that’s a different point!) Yet, even knowing that one of those theories is false, the scientific community continues to take both theories seriously, because we cannot simply dismiss a theory without having an alternative. Of course, scientists are actively looking for an alternative, and it is that search that is responsible for the development of string theory, loop quantum gravity, and other efforts to merge relativity with the quantum. But in the meantime, before we have such a theory, relativity and quantum mechanics remain the current, best theories available.
Of course, it’s one thing to point this out about the sciences, where explanatory knowledge is generally valued (despite the lure of empiricism, inductivism, Bayesianism, and other false theories of knowledge popular in some scientific communities). It is something else to apply the same view to other cultural pursuits. For one thing, other areas of knowledge often seem more susceptible to faddish trends, even when these trends offer only new ideas or data, rather than new explanations. (The trend I’m currently the most mystified by is Bayesian “rationality,” which is entirely non-explanatory and yet seems bewilderingly seductive to a lot of very smart people. But there are many others as well.) Another reason it’s difficult to apply this epistemological conservatism outside the sciences is that many of our current-best explanations in various non-scientific fields do look a bit politically conservative, which may make them less attractive in intellectual circles.
The question for this group, I suppose, is how much we participants want to see the “Re” as an invitation to re-assert existing good explanations that have not yet been rendered obsolete, but which are also losing steam in the wider culture—or, conversely, whether the “Re” is an spur to re-vise or even re-ject old-but-good explanations that no longer seem to hold cultural interest.
Let’s take the example of the ‘individual’ in biology defined in terms of genetic identity. We ‘know’ this is radically wrong (not just in the way that any scientific theory is – I like that you brought this nuance out Dorian) and yet it continues to be taught and propagated within and outside the academy. (A strong statement here is Margulis on serial endosymbiosis – though more exciting for me now is the view that the built environment (a termite mound functioning as a lung for the colony) is part of the ‘individual’). While I agree with your cautionary tone to some extent, I would hold that a lot of natural and social science is radically wrong in this sense.
I’d agree that a conversation pushing us forward on the concept of explanation, and explanatory knowledge would be most useful. . . it is something we have often said, but less often clarified or determined whether we have common understandings (or ones useful in their variations). Would be especially helpful to pick apart value of and methods for explanation versus description in knowledge production, and to get at whether we mean to explain what, or how. . . or, in the social sciences, why. Which gets very tricky.
There are cautions about getting too rigid on definition (thinking how economics pushed us toward explanation, then causal explanation, then explanation only if demonstrated with large scale randomized control trials . . which lost a lot along the way. But given the range of this group, I don’t anticipate that being the problem.
I wonder whether we could extend this conversation by recognizing that the “explanatory” concept of knowledge is fundamentally rational, and though this does not originate in Enlightenment thinking, it becomes dominant as scientific methods provide a paradigm of how knowledge works to offer descriptions of consistent laws of the behaviors/activities of the natural world (Descartes’s Method, Newton’s Physics, etc.). An alternative would be an “experiential” concept of knowledge that does not dismiss the general consistency of phenomena as represented within the explanatory paradigm, but recognizes alongside it the circumstantial specificity of knowledge as knowing. The tension between the two modes seems generative. One is always acknowledging the “yes, but” aspect of the simultaneously general and unique character of phenomena in the world as experienced in a human encounter (embodied, historically and culturally inflected, and so on).
I like Johanna’s “circumstantial specificity of knowledge as knowing” as a formulation complementary to the notion of “sitedness” from the Re:Enlightenment Report. It calls attention to the conditions that affect our efforts to forge better explanations without undermining the possibility of an explanation being better–that is, a better fit for the problem we are trying to explain.
Dorian’s question is tremendously important to Re:Enlightenment. Conserving as he describes it is a protocol–that is, an enabling constraint. Without the principle of conservation, advancement cannot take place. Knowledge zones in which explanations proliferate without constraint fail to produce better explanations.
Joining the question whether Re-Enlightenment may basically be a conservative project:
Going back to Kant “Was is Aufklarung,” enlightenment consists in
self reliance on one’s own reason rather than on authority. What are our
authorities? Whom do we wish to be enlightened?
What about the millions who rely on misinformation, thinking they are
relying on themselves? They really rely on authority: that of Trump or
the Party or the latent authority of the internet sites they chose to consult.
Books and newspapers also had latent authority, but the internet has
brought intensification through infinite repetition.
Evidence is crucial, then, for one can rely on reason but, if we think
of reason as a mode of computation, the results will be wrong if
the input is wrong. Of course, we may use reason to determine factuality.
I don’t know Kant well enough to say where he addresses the issue of
evidence
For Kant, Enlightenment arises from the individual investigation of the world, especially
religion. Enlightenment is Self centerered and applies to the individual, not to others except to the extent that the public exercise of reason can lead to the Enlightenment of
others. But, as Koselleck argued, this exercise of reason, at its extremes, can lead
to corrosive forms of critique and thus to destructive skepticism.
Have practices that allow us to determine facts though the analysis of evidence be hopelessly eroded? To the degree they have, the pervasive distrust of established modes of fact production that we see in millions of US citizens may be a new variant of the corrosion Koselleck defines.
Does the enlightenment in in Re-E become a mode of authority to be distrusted?
Reading through the provocations, I see Dorian and Helge as offering fundamentally different visions of the evolution of the online environment that Bill Blake describes in his. In Helge’s version, much as in Johanna’s idea of “experiential” knowledge above, the process becomes both the method and object of study. The urgency of new knowledge and the continual “NOW” that it labors under, which Helge describes, suggests a polyvalent, metamorphic approach to knowledge that is always reaching outside itself (trandisciplinarily, transinstitutionally) in the understanding of its own incompleteness. Like Dorian’s examples above, it fundamentally understands its own insufficiency as an explanation, but its (provisional) solution is a constant movement, a reaching towards a new now of knowledge, but which always risks forgetting even its immediate past. In the context of the online environment that Bill proposes, I imagine that this could look like a series of evolving experiments, or a continual layering and re-editing of a set of explanations. Either way, it proposes a dynamic, participatory environment that demands investment from us to keep up with its motion.
In Dorian’s vision, the conservatism of the good-enough explanation becomes a mode preservation. Again, there is the recognition that no explanation is sufficient, but, in this case, the reaction is one of skepticism and groundedness: instead of seeking after new explanations as a solution, this vision of knowledge creation rigorous tests all new innovations against the current explanations to ensure that they are fundamentally better explanations. Only then can they be allowed to supplant our current knowledge. Rather than Helge’s vision of a continually evolving now of experimental and experiential explanation, we have a productive resistance, wherein the accumulated failures of models to supplant the state of the art eventually builds up to a new explanation that is better than the current one. Here, the material realization of this within Bill’s proposed environment would focus on preservation rather than transformation: we might imagine the curatorial impulse of the museum or gallery wherein we can learn from an always-present past in order to build towards an incrementally better future. The web portal would let us always see the history of Re:E (and perhaps just the E) spread out, but would prioritize static curation rather than restless transformational knowledge. My chief concern in this case lies in the assessment of our explanations, whether preserved or fundamentally transforming: If we are choosing between a series of explanations, all of which, we acknowledge from the outset, are equally wrong, then how do we avoid picking the one that best fits our own worldview?
These two provocations, then seem, to me, to strike directly at the heart of the project of Re:E, whose dual impulse, as stated in all of our materials, is to look back and forward at the same time, to preserve and study as an method towards innovation. I have always regarded this tension as productive, but I wonder, in thinking through these alternatives, if the time is coming at which we will have to make a choice between cutting ties with good-enough explanations in the hope of transformative knowledge through innovation, or holding fast to the best available processes in the face of significant social transformation and its attendant pressures to change. Reading through these provocations with respect to Bill’s desire for a built environment, I find it difficult to imagine how we can build both.
All of that said, and I need to think through this further, I wonder if Lisa’s provocation might point the way out, as she draws our attention to the radically conservative social reaction to disruptive and transformative technologies.
I find Mark’s parsing of the Helga and Dorian alternatives extremely useful.
This is really clarifying, Mark. Let’s us it as a platform for our discussions later on today.
Much good provocation here about conservative, and conservation of knowledge. . . I’m feeling pulled toward the conserve stance, largely out of frustration with the work emerging post-pandemic in ed policy. We have always had short memories (Helge talks of a long now, but we have a short past, seldom more that a decade) in this area, and increasingly unlikely to remember the actions and interventions that were taken in the past, or the research and explanations developed about their effects. So looming on our horizon are summer schools, remediation, peer tutoring, the limitations of standardized tests, the value of recess/play, and how we learn to read. . . as if they were brand new ideas.
There are very real questions about how old is too old in social science, where some things certainly have changed . . .
When and which changes mean that old explanations can (or should) no longer stand as the best we’ve got? or when does the ‘we’ change enough to make them overly limiting or irrelevant?
But the big touchstone questions of Re:enlightenment, about past/present, and about preservation and dissemination of knowledge, not just development of new knowledge, seem more and more important in turbulent times.
Hi everyone! Thanks for the thoughtful comments. I’ll say more tomorrow, but for now, just a quick additional clarification, in case it helps as we think through these issues before tomorrow’s meeting.
I don’t mean that “conservatism” is about caution, much less about complacency with not-good explanations. When we have a better explanation, we should absolutely adopt it immediately. I am not advocating that we revert to Newtonian mechanics as our fundamental explanation just because we think Relativity is flawed. My point was that it, if we take explanations seriously, we should not give up an explanation in favor of a non-explanation. A great example is evolutionary theory. There’s that quip that in order to overturn evolutionary theory, we’d need to find a fossilized rabbit dating from before mammals existed. However, this would not overturn evolutionary theory, because we have no alternative good explanation for the emergence of life. In this scenario, we wouldn’t be sticking with evolutionary theory because we’re cautious, and we wouldn’t be sticking with it because we’re ok with a good-enough explanation; we’d be sticking with it because, if we take explanations seriously, it’s the only one we have–and meanwhile, we’d be searching constantly for a better one.
I’ll also just clarify that, in my original provocation, I was conceiving of “explanations” as being different from other kinds of ideas and hypotheses. I’m not advocating adherence to any old ideas.
I look forward to seeing you all tomorrow!
Buffon and others in c18 stressed that knowledge should concern itself
with “what” not with “why.” Which of these equals “explanation”? Both?
Interesting to see you comment below Dorian on evolutionary theory, which is an interesting case, being neither verifiable nor (obviously) experiential & thus provoking fideistic levels of attachment. Certain forms of knowledge are better than others at recognizing the provisionality of paradigms, but none are very good. I have been amused on an empirical level at the changing nature of explanations since the 1960s for giant mammal extinction post the last Ice Age & rather wonder if the proponents of the view that our ancestors hunted mammoths to extinction would like to try a field experiment with a flint-tipped arrow or two. No elephants with a rug on would be hurt in the making- that was sadly what the elephant gun was for- but I am not so sure about evolutionary biologists.
From the perspective of a training in computer science, the subtly different goals of research in the humanities and social sciences when compared with natural sciences and computer science is in my experience the most challenging component of collaboration. Newell and Simon wrote:
Computer science is an empirical discipline. […] Each new machine that is built is an experiment. Actually constructing the machine poses a question to nature; and we listen for the answer by observing the machine in operation and analyzing it by all analytical and measurement means available. Each new program that is built is an experiment. It poses a question to nature, and its behavior offers clues to an answer.
This corresponds with my experience of doing work in computer science. Systems are built and tested with data; often the system is at least implicitly designed as some kind of model of a real-world process, but actual explanations are often an afterthought. In social science, causal findings are most prized, and the key term for me here is modeling. We build a proposed model of the world, and then there is an empirical step where the model is compared to the real world system. The actual result of the research is an explanation (the mapping between the model and the real world system) but the empirical part is crucial. I’d be interested in discussing the relationship between models, evidence and explanations more.
I don’t believe that this way of arriving at explanations precludes a sited or specific way of knowing, but the sitedness of the communities conducting the research or creating the data surely sets the bounds of what explanations are proposed.
“descriptive” work in the social sciences seems to be undervalued; regarded as harmless but not of much value, while in the humanities, what might otherwise be called descriptive is sometimes viewed with more suspicion, as “description” may imply a neutral external perspective from which to describe. And maybe this brings us to back to question of our operating system or protocols: a broad and varied set of tools and protocols are our best hope for capturing the multiple specific ways of knowing.