![Sample our Humanities journals, sign in here to start your FREE access for 14 days]({"type":"image" , "src" : "/sda/5939/TOC-Subject-Sample-2021-Humanities.jpg"})
ABSTRACT
Despite its long period of dominance, the details of associationism as developed by the British Empiricists in the 18th and 19th centuries are often ignored or forgotten today. Perhaps as a result, modern understandings of Empiricist associationism are often oversimplified. In fact, there is no single core view that can be viewed as definitional, or even weaker, as characteristic, of the tradition. The actual views of associationists in this tradition are much more diverse than any such view would allow, even on fundamental aspects of the concept of association. This paper presents some of the most striking theoretical diversity and most significant discussions within the Empiricist associationist tradition. Recognizing this diversity, Empiricist associationism is best viewed as a project with shared questions and methods rather than a shared substantive theory of mind. This discussion can inform our understanding of the historical tradition, and also inform our current use of association and associationism. The concept of association is more flexible than it is often treated, and even its basic nature and function remain up for debate.
Acknowledgments
This paper began as part of an invited symposium on associationism to the 2020 Eastern APA, so thanks first to the organizer Kathryn Tabb, and participants Cameron Buckner and Talia Morag as well as attendees of the session. Thanks to the same as well as Michael Jacovides, Lauren Ashwell, David Cummiskey, Mark Okrent, Paul Schofield, and Susan Stark for helpful feedback on drafts. Thanks to the Bates Philosophy Department for ongoing support. Thanks, finally, to two helpful anonymous reviewers.
Disclosure statement
No potential conflict of interest was reported by the author(s).
Notes
1. The capital “E” indicates that Empiricist tradition, distinguishing it from empiricism as a more general orientation; I discuss the generally (lower-case) empiricist orientation of associationism below. Locke (Tabb, Citation2019) and James are contestable as members, and thus so are the precise endpoints of the tradition, but this won’t matter here.
2. For contrast, Anderson & Bower (Citation1973, Ch. 1) recognize that there is no single associationist theory, but still attempt to present a set of shared core associationist claims. I wish to rule out this strategy as a way of understanding the associationist tradition overall (which is admittedly not exactly their goal). The place to look is not in theoretical claims at all, but in the way they engaged in inquiry.
3. If one thinks, perhaps less dramatically, that it is more an exaggeration than a mirage, it won’t matter for the arguments to follow.
4. Readers skeptical of this claim can simply view χ-associationism as an invented foil that will motivate discussion of interesting and oft-ignored parts of the associationist tradition. It won’t matter much for the substance of the argument, and I am more interested in discussing the tradition than litigating perceptions of it.
5. One might reply that they are not “proper” associationists (e.g., Landes, Citation1926). But this is not my project. I work in the opposite direction: rather than starting with conception of associationism and seeing who fits, I start with the collection of authors as the tradition treated as associationist, and looking for a characterization of associationism that captures that collection.
6. There are other possible laws. Sir William Hamilton, a follower of Reid not generally considered an associationist, attempts to combine contiguity and similarity into the single law of association, which he calls “redintegration,” according to which “thoughts or mental activities, having once formed parts of the same total thought or mental activity, tend ever after immediately to suggest each other” (1877, lect 31). Bain was aware of this move by Hamilton, but treats redintegration as simply a version of contiguity (Bain, Citation1868, p. 327).
7. This is why association by contiguity remains an important explanation of learning today, in modified forms (Shanks, Citation2007).
8. This is perhaps vindicated by computational models that generate clusters of similar words by parsing their context (what words they tend to occur near) without needing pre-established categories or dimensions of similarity (Jones & Mewhort, Citation2007).
9. I take differences regarding reductionism and empiricism to be disagreements rather than debates per se: in both cases they are generally more interested in articulating differences with outside traditions than articulating differences within their own.
10. Including input/output associations might seem a trivial complication of χ-associationism. However, if one thinks through the implications, it forces a major step in the direction of liberalizing the concept of association in ways discussed above, which is why I left it out of my characterization of χ-associationism. It may even be that Mill missed the significance of this, but later authors did not.
11. For other recent attempts to draw modern conclusions from historical views, see, (Dacey, Citation2019a; Gokcekus, Citation2019).
Additional information
Notes on contributors
Mike Dacey
Mike Dacey is assistant Professor of Philosophy at Bates College in Maine, USA. He specializes in philosophy of cognitive science, especially history of psychology, modeling and explanation in psychology, and animal minds.