This post illustrates some points concerning how arguments were constructed in 18th century philosophy, which I made in my last post on the historical science-economics relationship.
Last summer I was staying over at someone’s house and happened to notice an old college copy of David Hume (1711-1776), I think An Enquiry Concerning the Principles of Morals (1751), sitting on a bookshelf. With a little downtime on my hands, I decided to have a quick skim. What struck me at the time was Hume’s use of historical events and poets’ observations as facts or phenomena that could be fit within a more systematized theory of human sentiments. I was going to write about that, but, going back, either I wasn’t reading the same thing, or Hume just doesn’t use the device as much as I thought (preferring more vague references to common experience and opinion). So, never mind that.
What did grab me on re-reading is Hume’s well-known argument against a reduction of human sentiment to self-interest, per Thomas Hobbes (1588-1679) among others. Hume framed his criticism in an interesting way:
An Epicurean or a Hobbist readily allows, that there is such a thing as a friendship in the world, without hypocrisy or disguise; though he may attempt, by a philosophical chymistry, to resolve the elements of this passion, if I may so speak, into those of another, and explain every affection to be self-love, twisted and moulded, by a particular turn of imagination, into a variety of appearances. But as the same turn of imagination prevails not in every man, nor gives the same direction to the original passion; this is sufficient even according to the selfish system to make the widest difference in human characters, and denominate one man virtuous and humane, another vicious and meanly interested.
The reductionism that Hume is talking about here is the supposition of basic elements or principles, which give rise to diverse and seemingly unrelated phenomena as they play out in a complex fashion. Much as Epicurean or Cartesian corpuscular motions, or (more in vogue in Hume’s time) Newtonian principles could give rise to seemingly unrelated physical phenomena, so, too, could every human impulse, whether clearly selfish or benevolent, be explained as various expressions of some self-interest.
Hume’s presentation of this sort of philosophical analysis of human emotions as “a philosophical chymistry” was not an idle polemical metaphor; it was a real methodological point. In the 18th century, the reduction of diversity in phenomena to the complex interplay of underlying elements and principles was an extremely common way of philosophizing. It was possible to work either backwards or forwards in this methodology, either starting from principles and logically surmising how they manifested themselves as phenomena, or taking phenomena, classifying them by some criteria, and then working out a logical sequence that would explain their classification.
“Chymistry” at that time was a major field where chemical transformation was used to try and determine the natural classifications of materials (acids, salts, earths, etc.), which substances were really primary, and what rules ultimately governed their interaction. This strategy would not lead to the promised principles in that or the next century, but it would lead to the chemistry of Lavoisier (1743-1794) and his followers.
Something very similar was playing out in botany. Here at Imperial, Staffan Müller-Wille recently gave an excellent talk to us about how Carl Linnaeus (1707-1778) went about developing his taxonomies of plants. Basically, Linnaeus understood there to be a major difference between an artificial classification (which is what he considered his famous system grouping plants according to their reproductive organs to be) and a natural one. Natural classification required much more intensive work, and as Müller-Wille shows, Linnaeus developed an elaborate cataloging method of grouping plants along a spectrum, which might loop back on itself (the last plant shows continuities with the first); but then individual plants might also cross-reference to points in other spectra as well. Untangling this process never produced a definitive result; it was an ongoing process of inquiry.
This sounded familiar, so in the Q&A I asked if Linnaeus constructed a natural philosophical system to try and explain natural classifications. Indeed he did: according to Linnaeus, there were only a few primitive species, which gave rise to other species through interbreeding. Apparently, today’s Linnaeus fans are a little embarrassed by his system, but the reduction of a system to the underlying principles that explained its manifestations was the thing for ambitious scholars to do, and Linnaeus was certainly an ambitious scholar. Did Linnaeus have any dealings with chemical experimenters? Yes, apparently he did.
This philosophical process is broadly described by what Michel Foucault identified as the “mathesis” of what he called the “grid” of the “Classical episteme” in The Order of Things (1966). There are of course problems with his blanket characterization of the thought of this period, but it certainly captures the prominence of the process of arranging and re-arranging phenomena in an attempt to capture their apparent interrelations, not only in natural science, but also apparently in human sciences such as linguistics. The “mathesis” was usually not explicitly mathematical, but it was an attempt to bring a logic to the patterns that repeated themselves in the world. (I’m pretty sure he viewed it as an analogue to his own “archaeology” of historical ideas.)
In establishing a theory of morals, Hume, also, was attempting to ground an understanding of social stability in the interplay of reason and various human sentiments. The question was: what should be the elements upon which the system was to be built? He allowed that some systems could be productively analyzed by reducing them to the interplay of simple, less-than-intuitive principles:
All attempts [to show benevolent affectations to be nothing but modifications of selfish affectations] have hitherto proved fruitless, and seem to have proceeded from that love of simplicity, which has been the source of much false reasoning in philosophy…. The case is not the same in this species of philosophy as in physics. Many an hypothesis in nature, contrary to first appearances, has been found, on more accurate scrutiny, solid and satisfactory. Instances of this kind are so frequent, that a judicious, as well as witty philosopher [Bernard de Fontenelle (1657-1757)] has ventured to affirm, if there be more than one way, in which any phenomenon may be produced, that there is a general presumption for its arising from the causes, which are least obvious and familiar.
Hume exempted moral sentiments from this idea, insisting that their varieties, causes and interplay could be most logically analyzed and ordered when enumerated at face value. This was manifestly not to say that the sentiments could not be ordered according to what sorts of situations evoked them, or that emotional ambiguity could not create confusion as to what emotions’ face values were: “Our predominant motive or intention is, indeed, frequently concealed from ourselves, when it is mingled or confounded with others, which the mind, from vanity or self-conceit, is desirous of supposing of greater force or influence…”
The point was that there was no hidden uniform metaphysical thought-structure giving rise to the various sentimental phenomena actually experienced: “…there is no instance, that a concealment of this nature has ever arisen from the abstruseness or intricacy of the motive…. We may as well imagine, that minute wheels and springs, like those of a watch, give motion to a loaded wagon, as account for the origin of passion from such abstruse reflections.”
The debate over the reduction of sentiments would not end with Hume, of course, but the terms of the debate would not remain stationary: the goals of theory and solutions to theoretical problems would morph from context to context. Much theory of circumscribed economic and political phenomena is now based on the abstraction from social reality of combinatorial calculations performed by self-interested “agents”, and much ink has been spilled over the epistemological legitimacy of making this sort of abstraction. Evolutionary biology analyzes the development of traits, including altruism, in terms of what selective advantages they offer. Friedrich Engels (1820-1895) would use the term “false consciousness” to describe sentiments arising from a psychological accommodation to the realities of social structures. Let’s not even get started on Freud.
These and other historical intellectual arguments can best be elucidated not in terms of a general epochal debate over what portrayals of “the mind” historical thinkers deemed properly scientific, but in terms of varying theoretical problems and varying contexts of standard intellectual practice. Conscientious historians need to delineate these contexts carefully.