Wednesday, September 27, 2017

Evening Note for Wednesday, September 27

Thought for the Evening: Indispensability Arguments for Causal Reasoning

Bertrand Russell, in his essay, "On the Notion of Cause" (found in Mysticism and Logic), famously argued that the notion of cause did not occur in advanced sciences, since it was not found as a term or relation in any of the best theories -- the role people attribute to causes is really taken by mathematical formulas. Russell himself would later pull back from this, and it has not generally been accepted; most philosophers of science today would, I think, regard it as false, although it perhaps depends to some degree on what one considers to count as causation. In any case, it has not fully died, either; I have occasionally run across people, including philosophers and physicists, making something like this claim. It is a sign of the fundamental problem that has plagued thinking about scientific inquiry for much of the past century and a half, namely, the obsession with theory. But the distinctive strengths of what we call scientific inquiry have never been found in theorizing; experimentation, broadly considered, and the reasoning based on it, is quite ineliminable. And here is the thing: there is no completely non-causal account of experimentation. We have no non-causal account of perception; we have no non-causal account of most kinds of measurement; we have no account of scientific inquiry in which experiments are not treated as effects of causes, namely, scientists; and, over and over, if you look at the reasoning used in experiments in order to draw conclusions from them, it is causal. There aren't even any semi-plausible-if-you-make-certain-assumptions rivals available; nobody has ever come up with any.

Suppose we take a simple experiment like this. A body is suspended from an arm of a properly constructed balance, with weights on the other arm in sufficient amount to balance out the body. We put another body some distance away and then electrify both bodies. If we see the body move, we add weights until it is back in balance, and take note of the weights we've added as a measurement. Even setting aside the fact that the very set-up of the experiment is entirely in causal terms (a balance we've constructed, a body we've suspended, a body we've placed, bodies we have electrified, weights we've added, measurements we've noted down), and even setting aside the fact that making sense of seeing the body move requires taking our visual experiences to be caused by the bodies in some way, to make sense of the measurements for the purpose of drawing the conclusion, we have to take the movement of the body on the balance as an effect of the other body, when they are both electrified. What could one conceivably do in order to strip out all this causal reasoning and replace it by noncausal reasoning, while keeping the conclusions about force completely intact? And the answer is that literally no one has any idea.

Thus just as some people want to run indispensability arguments for certain things in mathematics, based on the fact that you can't do science without them, so too one can run indispensability arguments for causal notions, on the same ground. Indeed, some of them are even easier to get off the ground, and come with less baggage than the indispensability arguments for mathematics, since there are at least simple scientific experiments you can do and reason about causally without having to bring in any mathematics at all.

Thus one could argue:

(A1) If something is indispensable for understanding our best experimental work and deriving theory-relevant conclusions from it, this is a good reason to accept it.
(A2) At least some causal notions are indispensable for understanding experiments, because (for instance) 'experiment' itself seems to be a causal notion, experiments use devices that can only be understood causally, we cannot draw conclusions from some experiments without causal reasoning.
(A3) Therefore we have good reason to accept causal notions.

Or we could argue instead:

(B1) Causal discourse is very effective as the language of experimentation.
(B2) There must be a reason for this success.
(B3) The simplest and most plausible reason is that at least much of this causal discourse is at least roughly true.
(B4) For causal discourse to be at least roughly true, some causal notions must be identifying something real.

Or we could argue:

(C1) In describing experiments so as to draw conclusions from them, scientists must operate on the assumption of the viability of at least some causal reasoning.
(C2) If we are justified in drawing conclusions this way, we must be justified in accepting causal reasoning.
(C3) To be justified in accepting causal reasoning, we must regard at least some causal notions as true.

Or we could argue:

(D1) There are genuinely causal explanations of experimental phenomena.
(D2) Some of these are our best explanations.
(D3) The best explanation for why they are the best is that they are at least approximately true.
(D4) For genuinely causal explanations to be at least approximately true, some causal notions must be identifying something real.

No doubt there are others one could use.

Various Links of Interest


* Giovanni Giorgini, Radical Plato: John Stuart Mill, George Grote, and the Revival of Plato in Nineteenth-Century England.

* Roger Scruton, As the left surges back, Marxism's bloody legacy is covered up

* Susan James, Mary Wollstonecraft's Conception of Rights

* Benjamin Schwarz, The New Elite's Silly Virtue-Signaling Consumption, a review of Elizabeth Currid-Halkett's The Sum of Small Things:

The cultural “products” that Currid-Halkett highlights as holding particular prestige for the educated elite—HBO dramas, TED Talks, podcasts, documentary films—are consistent with the gestural (one might say lazy) nature of elite intellectual activity. Consuming these products (or even reading Paul Krugman’s column) is entirely different from, say, wrestling with a thorny passage in the Book of Job or Das Kapital. Listening to a podcast or watching a TED Talk certainly exhibits and enhances cultural capital, but those are merely acts of passive consumption, rather than of intellectual and aesthetic engagement. Thus, for instance, Christian Lander has recognized the complacent and intellectually and politically stultifying character of so many undertakings that members of the elite tout as broadening: They “like feeling smart without doing work—two hours in a [movie] theater is easier than ten hours with a book.”

* Eleanor Parker, Stop teaching our children lazy anti-Catholic myths

* Robert Gooding-Williams, W. E. B. Du Bois, and Christopher Smeenk and George Ellis, Philosophy of Cosmology, and Erika Rummel, Desiderius Erasmus, at the SEP

* Pictures from Prime Numbers: The Trinity Hall Prime

Currently Reading


Susan Cooper, The Dark Is Rising
Edith Stein, Philosophy of Psychology and the Humanities
John of St. Thomas, Introduction to the Summa Theologiae of Thomas Aquinas
Maximos the Confessor, On Difficulties in the Church Fathers: The Ambigua, Volume I