I particularly hope to garner comments on this:

People claim to know things, but they mean different things by that. What is really irksome: People often assert a stronger claim for something being so, than they have any business doing. When these are people who have authority of one kind or another, that can become more than just irksome. So, this is an essay about evidence and belief and explanation. The bottom line: There’s far too much evidence and belief going around masquerading as scientific explanation.

Basic definitions first:

Evidence is information we have. It might be straightforward sensory intuition — I saw or smelled or tasted (whatever) something. It might be a collection of such events. I might take them as they appear to me; I might deal with them in some critical way. Such things are mostly qualitative experiences, but I can quantify them — this wine has a more fruity flavor than that; this student is clearly brighter than that. I can do fairly complex measurements in this way and plot the results and measure the amounts of change from one to another — the fruitiness of various wines across a spectrum, the acidity of those wines across a spectrum (we will leave the similar ways of dealing with pupils out of it — kindness, you know).

Some evidence goes beyond what I myself gather. Other people, we assume go through similar processes, and their evidence is available to me in one or another way. My parents made the evidence that constituted their experience available to me. Authors of books and papers do the same. I credit that information in different ways, depending on a variety of things; part of that assessment is based on the way the evidence is presented to me in explanation, about which more in a moment.

Belief is the integration and interpretation of evidence. I come to believe certain things about the constellation of events that is expressed in the evidence I have — in the elegant expression of Stephen Pepper: “world hypotheses”. That is, the sum of my evidence, taken together, underlies my understanding of my situation. This is fine, but can come acropper in at least two pretty obvious ways: On the one hand, I might very well extend my beliefs about the world in which I am situated beyond the limits of what my evidence supports. On the other hand, I may have explicit or implicit difficulties in accepting the evidence that others have compiled; this might arise from the way I have ordered my own evidence, anent the nature of those others. For example, if I don’t think you are as full able as I am, to experience and to compile your experience into coherent and meaningful evidence, I may dismiss that explanation. I may be right in doing so, or I may be very mistaken.

An extreme form of this kind of mistake shows up as totalitarianism. The beliefs I formulate, expressed as values, take on an objective character in a hierarchy that is conceived as a totality. Everything that fits, is just fine; everything that doesn’t fit, to the extent it doesn’t fit, slides toward unmitigated evil or non-existence. Those people who may be the source of such evidence as doesn’t fit, are also unmitigated evil, or not really extant and can be purged, one way or t’other.

Beliefs come in “flavors”. Some beliefs are carefully problematic; they express what one thinks is possibly so. Some beliefs are asserted as so, but the claim is not stringent. There are things of which one is certain; the claim is that such a belief is necessarily so. The last is the most demanding form, obviously. Equally, there is a real problem when merely problematic beliefs are presented as necessarily so.

To the extent we live together with others, and need to function together to mitigate the very real challenges that confront us, I have to explain my beliefs to you and you have to explain yours to me. One of us must persuade the other — a kind of violence involved in this, even if not the sort that entails bruises, or I have to ignore your view (or, you may choose to ignore mine — both of these sort of a “proto-totalitarianism”, perhaps, or we need to come to some agreement about the common world, a constellation of events that both of us recognize and to which both of us respond.

Explanation entails the test of belief, ultimately. I have a view of how things are, based on such evidence as I have. Does the evidence match the “real world”? It the image coherent? Does it allow me to project from what I believe to be the case to some future events with which I expect to be concerned?

It seems to me, to make this test, I have to be able to say with a high degree of certainty that things took just this form, for just this reason, through just this agency — and that in like circs., will do so again. I test this two ways: Can I set up the circs. to see if the predicted change takes place? If the change takes place, there is some indication I am justified in my beliefs (all the way back — that bit about coherence…); if the change fails, then to the extent it has done so, my beliefs are not robust and are subject to change. This is to say, the model of explanation is scientific explanation — causal (including all four Aristotelian causes), predictive and testable.

I would expect, in the normal course of events, that a system of beliefs would have to change, because the evidence available to me changes. If my beliefs don’t change — if I ignore the changing foundation of evidence — those beliefs will quickly diverge from their cognitive foundation. To the extent I act from such beliefs, my actions would be increasingly eccentric vis-à-vis whatever it is, external to myself, that is the source of the evidence appearing to me. That is, quite simply, a recipe for insanity.I started being concerned with this when I attended a so-ya-wanna-be-an-audiologist promotional “do” at Bellevue Hospital. Two CUNY dons (real ones, with research degrees) were introducing the new clinical doctorate (a non-research diplomate); the idea was, the new “evidence-based” approach to various kinds of medical practice, as well as general upgrading of the professional status of this particular specialty, justified the new degree. But, “evidence-based medicine”? “Now what the devil is that?” says I to myself.

A month or so later, my physician (nice lady, but giving up her cheap downtown practice to focus on her very posh Sutton Place clientèle) tells me she thinks I should start taking high blood-pressure medication. I do my due-diligence reading and come to the conclusion it can’t hurt, though the meaningful reports are actually very unclear. So I says to her, “OK, there are five or six categories of blood-pressure meds, and seven or eight players in each category. Which one and why?” Says she, “This one, because I like it.” Not “your high blood-pressure results from these factors and this med addresses them.”Nor is this really surprising; apparently, there is no good explanation of causative factors for high blood-pressure. Neither are there any good tests for more likely proximate causes of the problems for which high blood-pressure is commonly an antecedent condition. All there is, is lots and lots of case-histories, collated and counted and graphed in as neutral a way as can be. Evidence, it seems, is just this. What is puzzling, though: When this is presented to me, as explanation, it does not seem to stand up all that well to the tests I listed before:

  • In this instance, the most convincing evidence (a study done in Massachusetts, tracking a population over several decades) lists blood pressure as one of several causative factors, in conjunction with others. It is not at all clear — from the evidence as explained — which of these secondary causes is more significant, or whether any of them by itself is more commonly proximate. Moreover, the study looked at only one blood pressure related problem (heart attack).
  • Because these are understood only as secondary causes, there is no clear coherence among the data and with links to more primary causes which may not be proximate.
  • One accepts that elevated blood pressure is probably not a good thing but: It is not clear from the explanation that just controlling blood pressure symptoms will do more than stave off the eventual heart attack or stroke, eventuating from a primary cause not well understood.
  • The predictive character of such evidence is limited, at best. The explanation is not easily verified; criteria for falsification are almost too easy to discover and show to be the case.

This sort of evidence is not really convincing, and I am reduced to taking my physician’s advice on faith. Medicine is reduced to shamanism, or witchcraft. It has rituals and silly costumes and so on. There is even a litany (I have been hearing it chanted for the last 40 years).

[This is not a criticism of my physician, BTW.  She is, in fact, quite a decent person and, I think, a competent physician.  I think she does the appropriate professional reading; I think she hangs in the right circle to get the latest doctor-talk.  Nor does she necessarily make claims to science — wise, with me as a patient.  But the trappings of science, and sometimes the claims, are around, and they are just plain mistaken.

One commentator observed this might be a result of drug-manufacturer hype.  I think most competent medical professionals consider what is presented to them, but take it cum grano salis, along with other evidence.  That is a whole different question, finally; here my interest is the beliefs which some physicians adopt, explicitly or otherwise, convincing patients and perhaps in some way themselves, that they are being-scientific.  Something like that.]
More recently, I have had occasion to look at three ostensibly scientific papers from researchers on pigeon populations in cities. I inquired of their author about one of them because I was puzzled by the hiatus that seemed to obtain between assertion and supporting evidence; reading the papers convinced me that this was indeed the case, and that the problem was pervasive in the work coming out of this institute, part of an eminently respectable Swiss university.

[Keep in mind: I do favor pigeons — at least, the ones I visit with in the park by my house. First, they are remarkably friendly birdies. Even when I am not putting food right up in front of their little beaks, they will come and sit on my arm or hand, look me straight in the eye and just be, well, companionable. They seem to have very individual characters; they are all hungry, of course, but they respond to being fed, and to those of us who feed them, differently. They have a hard life, and yet it doesn’t seem to lead to ill-temper (as it can, for example, in dogs) — I have no way to understand this seeming stoicism. They sustain injuries of various kinds, and it is nothing short of amazing how they respond to being caught, held and having their wounds treated (far better than most children…). They are generally pretty clean (not a lot of pigeon poop — amazing considering that this group commonly includes 30 or more birds). While it’s summer and I am wearing short sleeves, the birds perching on my arms and hands can leave scratches, which have occasionally been just deep enough to draw blood; unlike cat scratches, which inevitably get a bit red and inflamed, these did not and quickly healed — prima facie evidence that these feral animals are cleaner than some folks’ house pets. I also confess, I am delighted that I don’t mind that I have the pleasure of these charming animals’ company, without having to clean cages or go for walkies at 6AM. All I need to do is show up with a cup or two of grain, bread crumbs, peanuts, lentils, unpopped popcorn every so often (interestingly, not even daily, and at variable times).]

The three papers from this one institute consisted of a literature survey, a report on pigeon-flock behavior and a report of diseases-pigeons-carry. Consider the paper, “Parasites from feral pigeons as a health hazard for humans”. The title suggests a positive causal relationship between pigeon parasites and human disease. The 8½ page paper (including an impressive list of references) describes the main tick and flea problems to which pigeons are prey, complete with appropriately unappetizing pictures of infested pigeons. In short, pigeons get ticks and fleas, just like dogs and cats. These make the pigeon sick, maybe.

But: The article observes that the fleas wash off in a shower and bath. While they can carry diseases (no surprise; so do other fleas — including some that are more prevalent and more likely to be picked up by human beings), the author of the paper cites only a limited number of instances of contact resulting in disease. Even allowing for under-reporting, the numbers appear statistically insignificant; people are more likely to die in a car-accident (even on the streets of Basel, one suspects).

Much the same is true of the ticks. The little blood-suckers that pigeons can reproduce only on a diet of pigeon blood; human beings in most cases will experience no problems; a few may have allergic reactions to the tick bite more or less the same way some people respond more strongly to mosquito bites. Stripped of the fancy lingo, a person might get a red inflamed bump and might even have secondary symptoms. Assuming reasonably prompt action (washing, perhaps the application of a topical cream of the same sort one would use for insect bites, based on hydrocortisone and maybe including an antihistamine — over-the-counter stuff even in the most regulated countries), problem instances are few, and allowing for underreporting, statistically insignificant.

The same author, in his paper, “Health hazards posed by feral pigeons”, observes, “Although feral pigeons pose sporadic health risks to humans, the risk is very low, even for humans involved in occupations that bring them into close contact with nesting sites.” People who are “immunocompromised” (that is, they are already sick in one way or another) may be more susceptible to pigeon-problems. The key word is “may”.

This is crummy science — better: it isn’t really science at all. The evidence presented does not extend adequately to form a foundation for the beliefs expressed in the paper titles and conclusions. The presentation of that evidence — the explanatory parts — fails to get beyond the merely problematic. But the title and concluding statements are couched in apodictic language. In short: Academic smoke and mirrors.

So, the tale is quickly told:I think a besetting sin at the onset of Post-Modernity is that we are holding as necessary beliefs which extend far beyond any evidence we have, or even, may have. The mistakes would be comical, were the consequences not so grave.

Advertisements