A Theory of Rationalization
The U.C. Santa Cruz philosopher Jon Ellis and I are collaborating on a paper on rationalization in the pejorative sense of the term. I'm trying to convince Jon to accept the following four-clause definition of rationalization:
A person -- whom, following long philosophical tradition, we dub S -- rationalizes some claim or proposition P if and only if all of the following four conditions hold:
1. S believes that P.
2. S attempts to explicitly justify her belief that P, in order to make her belief appear rational, either to herself or others.
3. In doing 2, S comes to accept one or more justifications for P as the rational grounds of her belief.
4. The causes of S's belief that P are very different from the rational grounds offered in 3.
Some cases:
Newspaper. At the newsstand, the man selling papers accidentally gives Estefania [see here for my name choice decision procedure] a $20 bill in change instead of a $1 bill. Estefania notices the error right away. Her first reaction is to think she got lucky and doesn't need to point out the error. She thinks to herself, "What a fool! If he can't hand out correct change, he shouldn't be selling newspapers." Walking away, she thinks, "And anyway, a couple of times last week when I got a newspaper from him it was wet. I've been overpaying for his product, so this turnabout is fair. Plus, I'm sure almost everyone just keeps incorrect change when it's in their favor. That's just the way the game works." If Estefania had seen someone else receive incorrect change, she would not have reasoned in this way. She would have thought it plainly wrong for the person to keep it.
Wedding Toast. Adrian gives a wedding toast where she tells an embarrassing story about her friend Bryan. Adrian doesn’t think she crossed the line. Yes, the story was embarrassing, but not impermissible as a wedding toast. Shortly afterward, Bryan pulls Adrian aside and says he can't believe Adrian told that story. A couple of months before, Bryan had specifically asked that her not to bring that story up, and Adrian had promised not to mention it. Adrian had forgotten that promise when preparing her toast, but she remembers it now that she has been reminded. She reacts defensively, thinking: "Embarrassing the groom is what you're supposed to do at wedding toasts. Bryan is just being too uptight. Although the story was embarrassing, it also shows a good side of Bryan. And being embarrassed like this in front of family and friends is just the kind of thing Bryan needs to help him be more relaxed and comfortable in the future." It is only because Adrian doesn't want to see herself as having done something wrong that she finds this line of reasoning attractive.
The Kant-Hater. Kant's Groundwork for the Metaphysics of Morals -- a famously difficult text -- has been assigned for a graduate seminar in philosophy. Ainsley, a student in that seminar, hates Kant's opaque writing style and the authoritarian tone he thinks he detects in Kant. He doesn't fully understand the text -- who does? -- or the critical literature on it. But the first critical treatment that he happens upon is harsh, condemning most of the central arguments in the text. Because he loathes Kant's writing style, Ainsley immediately embraces that critical treatment and now deploys it to justify his rejection of Kant's views. More sympathetic treatments of Kant, which he later encounters, leave him cold and unwilling to modify his position.
The Racist Philosopher. A 19th century slave-owner, Philip, goes to university and eventually becomes a philosophy professor. Throughout his education, Philip is exposed to ethical arguments against slave-ownership, but he is never convinced by them. He always has a ready defense. That defense changes over time as his education proceeds and his thinking becomes more sophisticated. What remains constant is not any particular justification Philip offers for the ethical permissibility of slave-ownership but rather only his commitment to its permissibility.
These cases might be fleshed out with further plausible details, but on a natural understanding of them the primary causes of the protagonists' beliefs are not the justifications that they (sincerely) endorse for those beliefs -- rather, it's that they want to keep the $20, want not to have wronged a close friend at his wedding, dislike Kant's writing style, have a selfish or culturally-ingrained sense of the permissibility of slave-ownership. It is this disconnection between the epistemic grounds that S employs to defend the rationality of believing P and the psychological grounds that actually drive S's belief that P that is the essence of rationalization in the intended sense of the term.
The condition about which Jon has expressed the most concern is Condition 4: "The causes of S's belief that P are very different from the rational grounds offered in 3." I admit there's something that seems kind of fuzzy or slippery about this condition as currently formulated.
One concern: The causal story behind most beliefs is going to be very complicated, so talk about "the" causes risks sweeping in too much (all the causal history) or too little (just one or two things that we might choose because salient in the context). I'm not sure how to avoid this problem. Alternatives like "the explanation of S's belief" or "the real reason S believes" seem to have the same problems and possibly to invite other problems as well.
Another concern: It's not clear what it is for the causes to be "very different" from the rational grounds that S offers. I hope that it's clear enough in the cases above. Here are some reasons to avoid saying, more simply, that the justifications S offers for P are not among the causes of S's belief that P. First, it seems typical of rationalization that once one finds some putative rational grounds for one's belief, those putative grounds have some causal power in sustaining the belief in the future. Second, if one simply couldn't find anything even vaguely plausible in support of P, one might have given up on P -- so the availability of some superficially plausible justifications probably often plays some secondary causal role in sustaining beliefs that primarily arise from other causes. Third, sometimes one's grounds aren't exactly what one says they are, but close enough -- for example, your putative grounds might be your memory that Isaura said it yesterday, while really it was her husband Jeffrey who said it and what's really effective is your memory that somebody trustworthy said it. When the grounds are approximately what you say they are, it's not rationalization.
So the phrase "the causes... are very different" is meant to capture the idea that if you looked at the whole causal picture, you'd say that neither the putative justifications nor close neighbors of them are playing a major role, or the role you might normatively hope for or expect, in causing or causally sustaining S's belief, even as she is citing them as her justifications.
What do you think? Is this a useful way to conceptualize "rationalization"? Although I don't think we need to hew precisely to pre-theoretical folk intuition, would this account imply any particularly jarring violations of intuition about cases of "rationalization"?
I'd also be happy for reading recommendations -- particularly relevant philosophical accounts or psychological results.
Our ultimate aim is to think about the role of rationalization in moral self-evaluation and in the adoption of philosophical positions. If rationalization is common in such cases, what are the epistemic consequences for moral self-knowledge and for metaphilosophy?
-------------------------------------------
For related posts, see What Is "Rationalization?" (Feb. 12, 2007), and Susanna Siegel's series of blog posts on this topic at the Brains blog last year.