FRONT PAGE / POSTS
The Abductive Method
by Joshua Glenn| Boston, USA
Thursday, 2 December 2010
tags: americas, culture, europe, making sense, semiotics
I've enjoyed watching the first three episodes of BBC's Sherlock (aired this fall in the US; starring Benedict Cumberbatch and the brilliant Martin Freeman as Dr. Watson), particularly because doing so prompted me to revisit the notes I took, then put aside, when reading The Sign of Three (Indiana University, 1983), a collection of essays about Arthur Conan Doyle's character and the semiotician C.S. Peirce — or more precisely, Peirce's theory of a little-understood mode of reasoning he named "abduction."
Deduction, according to Peirce, proceeds from rule/guess (e.g., "All the beans from this bag are white") to case ("These beans are from this bag") to result/observation ("These beans are white"), whereas induction proceeds rather more tediously — from case to result to rule. I say "tediously" because a guess based upon both case and result/observation is a safe, habitual guess; detectives, who form hypotheses and then test them against the case (evidence), are more romantic figures. However, the authors included in the book mentioned above — including Thomas A. Sebeok, Marcello Truzzi, Carlo Ginzburg, Jaakko Hintikka, and Umberto Eco — pooh-pooh Holmes' vaunted powers of deduction. Though Doyle's stories do a terrific job explaining how deduction ought to work, Holmes' skill at solving crimes is due, they claim, to a brilliant abductive ability — i.e., the ability to proceed, swiftly and with unerring accuracy, from rule/guess to result/observation to case.
Like the deductive reasoner, the abductive reasoner begins with a rule/guess: e.g., "All the beans from this bag are white." By comparing a result/observation ("These beans are white") against the rule, though, the abductive reasoner doesn't seek to test the validity of her hypothesis, but instead to detect any deviations from it. Which shouldn't exist!
Abduction is something that all of us do, claims Peirce; in fact, it's a hard-wired survival mechanism. However, he and Sebeok, et al., agree that some of us are particularly adept at abductive reasoning. Some of us see and remember more, so we're superior at formulating rules/guesses; and then, when we compare a result/observation against one of our rules/guesses, we do so ultra-efficiently — in an almost holographic fashion. Those of us thus skilled at detecting deviations from law-like hypotheses are therefore able to see the reason why "like a flash," claims Peirce. What's more, the act of abduction is in such cases accompanied by a "peculiar musical emotion," a thrill.
Sounds like Sherlock Holmes — his monographs on cigarette butts or corpses' bruises, his lightning-fast insights, even his boredom and mood swings. The new BBC adaptation dramatizes Holmes' holographic ability to compare a result/observation against one of his rules/guesses by causing words, patterns, and symbols to hover in the air before his face [shown above]. He's viewing the evidence not empirically, we're given to understand, but from the perspective of his own constructed universe: if Holmes' hunches are always correct, it's only because this is fiction.
Though he insists that his method is a strictly deductive one, at various points in the Holmes canon, Conan Doyle's detective advocates the use of "imagination," "intuition," and "speculation." This explains why his so-called deductions so often lead Holmes to make revelations which appear almost magical; and this is why Holmes despairs of colorless, boring cases. He's an obsessive, quasi-apophenic pattern-maker. When he finds a flaw in the pattern, he's thrilled; when he doesn't, he's bored. He's an obsessive-compulsive overjoyed and outraged to find reality out of order.
Holmes sounds, in this analysis, like a semionaut — i.e., a prodigy able to draw expertly and productively upon phenomenological knowledge when "reading" various signs. Yes, Holmes is a semionaut. However, I'm not always impressed with the immutable laws of nature and society of which Holmes has convinced himself. Though he says "I make a point of never having any prejudices" ("The Reigate Puzzle"), not a few of Holmes' rules — about the habits of women, say, or foreigners — sound, to the contemporary reader, like prejudices. In the third episode of Sherlock, when Cumberbatch takes one look at a woman's boyfriend and says, simply, "Gay" — same thing, right?
I'm not saying that Holmes is merely a brilliant bigot, like (say) G.K. Chesterton's fun detective character, Father Brown, who solves crimes thanks to his hilarious Catholic prejudices against atheists, legalists, and Presbyterians. But he's uncannily similar to a brilliant bigot. It's mysterious!
The game is afoot.
17 December 2010 at 2:25 pm
Joshua Glenn says:
It’s a terrific question, Malcolm. Abduction is what we semiologists and semionauts do… but (a) the word sounds creepy [Pierce offers several other terms for the same mode of reasoning/perceiving, so we could figure out a better one]; and (b) as you note in your comment about subjective opinion vs. “science” the method wouldn’t sound scientific to clients — which is a good thing for guru-type semiologists, a bad thing for semiologists who don’t market (or perceive) themselves as gurus.
Of course, the answer is: Sherlock Holmes! Who insists over and over again that he’s a Mr. Spock-type living computer, super-rational and scientific, a master of deduction and close observation… all the while behaving like a guru — e.g., his parlor trick where he tells a client intimate details about herself based on a quick glance.
Seriously, though: semiotic brand analysis which integrates itself firmly and in a mutually beneficial way with consumer insight research combines “aha”/abduction with something closer to scientific-style deduction or induction, the result of that merger being something of which Peirce could approve. Whether or not the result is “true” is a more difficult question — but it’s highly persuasive.
16 December 2010 at 7:58 pm
Malcolm Evans says:
I’d be interested if anyone’s ever packaged up abduction as part of a commercial semiotic offer. It’s an interesting challenge how best to communicate it in terms of benefits for brands in language clients would understand and respond to. Would one mention ‘abduction’ at all or go via another route? The correction of bias you describe as coming after that “first state of inquiry”, Josh, feels like potentially a powerful offer to follow up with in the context of commercially applied semiotics. It also feels like there’s some potential via Peirce and abduction for a refreshing methodological transparency compared with an approach based more on semiology (even if it’s branded ‘semiotics’) which often seems poised uncomfortably between the rhetoric of what Saussure dubbed a ‘science’ and a practice of apparently deeper than normal interpretation which invites the question “why isn’t this just subjective opinion?” But maybe abduction doesn’t quite answer that one either. I know there are highly plausible answers to that (informed opinion, grounded hypotheses etc) and I can hear my own mind and those of colleagues revving up to address the issue even as I write. But it’s always a stretch to get to those answers. Something about them feels second order, semantically delegated upstairs, a little furtive in spite of themselves. And they don’t sit as comfortably in the realm of everyday good sense as the question does. You can imagine the question (but not the answer) as articulated by Bob Dylan, Jay Z or Oprah. My (abductive) gut feel is that abduction delivered via an everyday metaphor or metonymy may help get that answer into a more intuitively persuasive register.
14 December 2010 at 3:56 pm
Joshua Glenn says:
>We shouldn’t be too averse to stereotyping as
>part of that process. That may be fundamental
>to perception, especially if it’s perception
>linked in any way to purposive action. If you
>don’t have certain preconceptions about the
>road, the crossing, the red and green flashing
>people and the beep beep beep noises you’re
>going to remain wide open to a richly nuanced
>cornucopia of sensory impressions and never
>cross the road.
Agreed!
>If you manage to avoid that
>perceptual plenitude and paralysis of will it
>doesn’t make you an anti-red man bigot or
>anything like that.
True — except for when it actually does make you a bigot. In the episode of “Sherlock” I mentioned, when Holmes takes one look at a woman’s boyfriend and says “gay,” we assume he’s right, because his abductive powers have never failed yet. However, it turns out later in the episode that the boyfriend was actually MORIARTY! Pretending to be gay! Moriarty took advantage of Holmes’ faith in the near-infallibility of his own powers of creative abduction — which he seems to believe is actually deduction. Moriarty found Holmes’ blind spot: Even Peirce, the great champion of creative abduction, warns that in any honest investigation, the certitude offered by abduction mustn’t be allowed to exclude fallibilism (Peirce: “the doctrine that our knowledge is never absolute but always swims, as it were, in a continuum of uncertainty and undeterminacy”).
Peirce insisted that although abduction is a terrific “first state of inquiry,” the investigation’s second state must begin not with scrutiny of the phenomena but of the hypothesis. In order to correct for bias.
14 December 2010 at 11:42 am
Chris says:
Totally agree with all this. Abduction is at the heart of the semiotic method.
7 December 2010 at 6:42 am
TH says:
There is a whole branch of medical semiotics that I’m interested in exploring. I’m often struck by how many professions share the fundamental challege of problem-solving. This can be viewed/approached as semiotic. And it can be telescoped/microscoped back-and-forth in scale from something as broad as a universal human quest for meaning to a mechanic trying to fix the gearbox. (Maybe the gearbox lets us do the scale-shifting.)
fMRI info can be found in the medical literature. There are some subscription-only references (http://cdp.sagepub.com/content/18/4/210.full.pdf+html), but here are some brief summaries:
http://www.wisconsinmedicalsociety.org/savant_syndrome/whats_new/archive/2005
(go about halfway down)
and:
http://www.timesonline.co.uk/tol/news/uk/health/article822534.ece
There is a danger, of course, in taking medical research as absolute. Combined with prejudice, people come up with things like the Third Reich (and then blame philosophers for them!) But there is, I agree, Malcom, a role for prejudice. We have to discriminate between the white, green, and red beans or else all the beans are the same, nothing means anything, and we all die of starvation, poisonous plants, or traffic accidents.
We seem to need some place on which to stand in order to perceive and take action on something else.
6 December 2010 at 7:23 pm
Joshua Glenn says:
Right — I’m not anti-prejudice! Far from it. But in the case of Sherlock Holmes or Father Brown, where you’ve got a detective solving a case because he knows exactly how a Negro or Presbyterian would behave in certain circumstances — well, it’s unseemly to say the least.
Great description of how we do semiotic analysis, Malcolm. TH, I’m fascinated by your mention of fMRI and the “eureka” moment. I’d love to know more about this…
6 December 2010 at 4:04 pm
Malcolm Evans says:
There’s no doubt that abduction is key to innovation and to the activity of successful semiotic analysis. You can’t understand the significance of the parts until you’ve grasped the whole and you can’t see the whole until you’ve aggregated the parts – so you have to start from both directions at once, formulating hypotheses and bouncing them off each other, looking for clues, doing informed guesswork. By telling yourself stories, however preposterous, about the limited data you’ve got you’ve fabricated a context which opens up other potential vistas on the data and other narrative possibilities. We shouldn’t be too averse to stereotyping as part of that process. That may be fundamental to perception, especially if it’s perception linked in any way to purposive action. If you don’t have certain preconceptions about the road, the crossing, the red and green flashing people and the beep beep beep noises you’re going to remain wide open to a richly nuanced cornucopia of sensory impressions and never cross the road. If you manage to avoid that perceptual plenitude and paralysis of will it doesn’t make you an anti-red man bigot or anything like that. You can even think the red man’s a monkey if you’re crazy enough and nobody’s going to mind as long as you don’t try to put him in a cage. If you introduce the red and green men to the pile of beans you’ve got poetry (especially if they’re big eyed beans from Venus), which is just another form of abduction. The fact that Peirce and Sebeok have abduction and that semiotics in their book is rooted in nature not just in culture (which is, after all, just a part of nature) is drawing me irresistibly in their direction after years and years of monkeying around with the French and their semiology. Much as I applaud the French for their cheese, general cuisine, aversion to psychotic shock-and-awe style behaviour, and stereotypical je ne sais quoi.
5 December 2010 at 9:21 am
TH says:
Most people investigating something (a system, problem, condition) typically use some combination of inductive, abductive, and deductive reasoning.
Scientists use the scientific method to arrive at some sense of reality. (What’s the Arthur C. Clarke quote: “Reality is that which, when you stop believing in it, doesn’t go away”.) First comes inductive reasoning: we look at cases (random beans from the bag), find patterns (all the beans we’ve gotten so far are white), then determine a “rule” (all the beans from this bag are white). The “rule” then becomes a theory that can be applied prospectively. Additionally, the “rule” is tested by subsequent hypotheses (thought note with Peirce, “hypothesis” is based not on probability, but aberrancy).
This approach works well for cases in which no rule is known at the outset. Also for systems that can be tested repeatedly.
The more the rules of a system are established, the more deductive reasoning comes to the fore.
There are many areas in which pre-existing “rules,” or at least their application, are quite nebulous. Medicine is one such area. In medical diagnosis, we are called upon to analyze a result – beans lie on the table. We then need to use the right rule(s) to understand the specific case. Pathophysiology follows some “rules,” yet the manifestation of these rules can vary as widely and wildly as one human being from another. Nor is there always a second chance to get it right. Medicine is one area in which we see a lot of abductive reasoning.
Detectivework is another such area. And it is no wonder that Sherlock Holmes’ creator, Sir Arthur Conan Doyle, was a physician. The eponymous Dr. House of television is perhaps an anti-Holmes in his use of abductive reasoning. He misdiagnoses all the time, performs unnecessary procedures (up the wazoo, so to speak). (He also walks with his cane on the wrong side.) But his is a different sort of fiction.
Abductive reasoning is heavily reliant upon (and certainly overlaps) insight-based reasoning. Both pivot upon an “aha” or “eureka” moment. In research to map the “aha moment” in the brain, functional magnetic resonance imaging has shown solving a problem through an “aha moment” engages quite a different pattern of neuronal activity than slogging through deductive reasoning.
In some studies, there has been a flare of neuro-electrical activity in an unique part of the brain just preceding the “eureka”. It has been postulated that this difference has involved the “breaking of mental sets” with an “aha moment” – perhaps literally thinking outside the box. One might also invoke Arthur Koestler’s “Act of Creation” to postulate currents flowing at perhaps a subconscious level that then suddenly converge in a moment of realization/creation. Perhaps it is even akin to a seizure. Fyodor Dostoyevsky would say his seizures, while not contributory per se to his storying brilliance, at least to his knowledge, were among the most blissful states.
At any rate, the image of humans performing fMRIs in search of data on “aha moments” is a great metaphor for the human quest for understanding – the brain reaches for an “aha moment” in understanding the “aha moment”.