Inspired by Plato onwards, theories of cosmic, physical, and moral sympathy (συμπάθεια--'fellow feeling') were developed in a variety of contexts (e.g., Galenic medicine, Stoic metaphysics, magnetism, moral psychology, magic, etc.). For all its variety, in most thinkers and traditions the very possibility of sympathy presupposes that sympathy takes place among things that are in one sense or another alike (sometimes within a single being/unity/organism) to be contrasted with the antipathy (ἀντιπάθεια) of un-alikes. (Here I just flag the non-trivial moral issues this raises for ethical theories that rely on sympathy/empathy.) Let's call this condition of the possibility of sympathy, "The Likeness Principle" (or TLP). I learned the significance of the TLP in Plotinian and Stoic thought from Eyjólfur Emilsson, René Brouwer.
Recently, Stewart Duncan has been blogging about sympathy in various contexts (e.g., Margaret Cavendish) including a nice treatment of an important response by Henry More to Marin Mersenne. The context is a debate over what counts as a proper explanation of sympathetic resonance of vibrating strings. [Among historians of science, the best known 17th century discussions (unmentioned by Duncan) occurs in Galileo's treatment of resonating pendula at the end of the first day of the The Discourses and Mathematical Demonstrations Relating to Two New Sciences (Galileo avoids using 'sympathy') or Gilbert's work on the magnet and action at a distance (recall my post).]
All which things do clearly shew, that pure corporeal causes cannot produce this effect: and that therefore we must suppose, that both the strings are united with some one incorporeal Being, which has a different Unity and Activity from Matter, but yet Sympathy therewith; which affecting this immaterial Being, makes it affect the Matter in the same manner in another place, where it does symbolize with that other in some predisposition or qualification, as these two strings doe in being tuned Unisons to one another (More 1659, 452-3).
Strikingly enough, for More sympathy is possible for two things (matter and incorporeal being) that are largely unalike and, thus, almost seem to violate The Likeness Principle. I write "largely" and "almost" because the Platonizing More stipulates some (extremely thin) likeness: "some predisposition or qualification" (whatever it may "symbolized" to be). So, unless one is a Cartesian substance dualist, sympathetic mind-body interaction need not violate TLP. (Given that Descartes had rejected sympathy as occult at Principles 4.187, appealing to it is not a live option for him when pressed to explain mind-body interaction.)
Yet, recently the distinguished Renaissance scholar, Ann Moyer, called my attention to the following passage in a chapter she is contributing to a volume that I am editing on sympathy in the Oxford Philosophical Concepts series (edited by Christia Mercer):
"And euen naturall Philosophy teacheth this, that betwéene the soule and the body, there is a certaine sympathy or knitting of affection: for who seeth not that in melancholy bodyes the mynde is heauy and solitary, in sanguine bodies mery and lyght, &c."--Richard Cavendish The Image of Nature and Grace (1571).
Here, there is sympathy between mind and body. Of course, the possibility of sympathy may (as Cavendish seems to imply) be a consequence of the soul being embodied (in mortals) and, then, the TLC may be saved (because sympathy is taking place within a single being/unity/organism). In Cavendish there is nothing mysterious about sympathy between mind and body (it is justified with an appeal to "natural philosophy"). However, as Stewart Duncan reminded me, within a century, the violation of the TLC is evidence for God's (providential) intervention:
"For this Sympathetical Union of a Rational Soul with Matter, so as to produce a Vital communication between them, is an arbitrary institution of the Divine Wisdom: there is no reason nor foundation in the separate natures of either substance, why any Motion in the Body should produce any Sensation at all in the Soul; or why This motion should produce That particular Sensation, rather than any other." Richard Bentley (1692-3) The Folly and unreasonableness of atheism.
As Stewart Duncan points out, Bentley here deploys a version of Lockean super-addition. (I don't think Locke ever appeals to sympathy in doing so, but I may be wrong.)
Bentley's piece is from his famous Boyle lectures (the first). In fact, we know that in turning these Boyle's lecture from sermon to publication he induced Newton into a famous correspondence over the nature of action at a distance. The interpretation of that Newton-Bentley echange is still controversial (see my treatment of these matters--the only thing not in doubt is that Newton explicitly rules out that Newtonian attraction should be identified with Epicurean approach to gravity). Now, action at a distance does not violate the TLP. For in Newton, all matter is fundamentally alike in having mass. It is an open (and much debated) question if Newton was committed to any other principle that would have blocked such action at a distance, but that needs to be explored elsewhere (recall).