#50: Cameras, Hypotheses, and Politics

The Leica 35mm still camera was introduced to the world in 1913. Designed by Oskar Barnack and produced by Leitz in Wetzlar, Germany, the camera was simplicity itself. It was, in effect, no more than a tiny metal box holding a strip of the then readily available 35mm movie film; a box with a lens, a simple cloth shutter, a mechanism for advancing the film one frame at a time, and a primitive viewfinder tacked to the top. The lens was a small (f. 3.5) uncoated Elmar (the image doesn't show the Elmar).

While the simplicity of the design accounted for the camera’s rapid acceptance, the simplicity was also experienced by many as an annoying handicap. Leitz addressed the little pocket camera’s shortcomings by designing and producing an ever increasing number of “adapters,” small devices that could be attached to the camera body by various tiny screws. For example, while the original Leica had to be focused by estimating the distance to the object, one could also purchase a range-finder device that could be mounted on the camera. Over time, Leica spawned an awesome number of such auxiliary attachments, some of them being attachments for attachments, until a complete photographer would venture out into nature with his trusty tiny Leica and a large bag filled with all the many expensive little screw-on devices.

The point to notice is the absence of any voices in Wetzlar telling the marketers and designers that they should either abandon their original design for one which accommodated the photographer’s needs in its initial conception or one that integrated the desired functions more naturally into the Leica design. The Wetzlar approach was a case study in the notion of over-engineering.

Over-engineering is the result of an invincible unwillingness to abandon a design whose potential has been exhausted. The German engineers who worked on the Leica and a number of its successors were apparently fatally in the grip of the over-engineering impulse. Psychologically, I suppose this impulse can be described as a kind of “rigidity.” Today, there are people who would describe them as being unable to “think outside the box”; it would be more apt to say that they had a compulsion to add things to the box.

It was not until 1933-36 that a fundamentally new design appeared on the market and, to be fair, it appeared in Germany in the form of the Exakta single lens reflex. While it was a definite success, acquiring a very large following, it still retained, in my opinion, the over-engineering impulse of the Leica. Ohmygawd, did they ever produce screw-on supplementary gadgets for this thing! I was not, however, immune to its allure; I still own a late model Exakta VX1000!

In 1951 the single lens reflex seemed to move one notch higher, and this time the improvement did not come from Germany. The new design was that of the Japanese Asahi Pentax single lens reflex. While the Exakta offered through-the-lens focusing and viewing, which made access to new lenses of varying focal lengths much more accessible, it was still limited to waist-level viewing. The Pentax took the Exakta concept and added a five-faced prism (the "pentaprism") that made eye-level viewing and focusing possible. A huge improvement. The Germans, of course, now also followed suit, but we didn’t see Zeiss’s offering, the superbly machined Contarex, until 1958. I should add that the Contarex weighed probably twice as much as the Pentax. In all of this, I also add as a personal opinion, that the German lenses were throughout unquestionably superior to the Japanese ones. The Germans made really good glass! The only lens I’ve ever come across that could compete with the top Leitz or Zeiss lenses was the ridiculously expensive, rather rare, and comparatively obscure Switar range of apochromatic lenses on the Swiss Alpa. Absolutely beautiful glass!

The single-lens reflex allowed for a number of significant improvements over the original Barnack design, retaining only the 35mm film format. Focusing, and soon after, metering took place directly through the taking lens, eliminating the two most important “gadgets” added to the Leica box: the range-finder and the external light meter.

For most shooting, this was simply a much better design and the Japanese went into it with complete commitment and enthusiasm. The Germans, not so much. They just couldn’t abandon their old good idea for a new, much better idea.

Why, I hear you ask, are you bothering me with such arcane historical trivia? There are two reasons for this. The first is that the patterns of technological development are always inherently interesting; the second is that the compulsion to over-engineer is not limited to the world of machine production.

The terminology is different, but the compulsion is the same. In the softer sciences and some “sciences” that are not sciences at all (such as political “science”), we come across the notion of the “ad hoc” hypothesis. This notion is a kind of abstract equivalent of over-engineering.

The ad hoc hypothesis is a supplementary assumption added to a primary or main hypothesis in order to keep it from failing. Say that your primary hypothesis is A, but that you encounter evidence E that is incompatible with A. Your initial and healthy response is likely that A should be rejected because of E. But it does not have to be. We can modify A, for example, by restricting the cases to which it applies, thus avoiding conflict with E. Or, we could add a supplementary theory that “explains” in some way exactly why E should not be taken to count against A. In general, we tend to count as ad hoc hypotheses only such additions as entail changes to the theory included in the original main hypothesis.

Neither supplementary engineering nor supplementary theorizing are necessarily objectionable. Just adding a range-finder to the first Leica did not count as over-engineering, just a reasonable complement for a nice piece of equipment. Similarly, making a small theoretical modification of, say, a physical hypothesis may be nothing more than an appropriate adjustment to new data. The problem is that there eventually comes a point when both are clearly no longer reasonable adjustments, but that there seems to be no clear dividing line separating the reasonable from the compulsively crazy.

In both cases, the symptom is usually an extreme complexity in the final product. When the original inspiration or “concept” of the Leica was no longer discernible under the mass of screwed-on supplementary junk, it was then that an innocent bystander would have said: it’s time for a new idea.

It was the simplicity of the original Barnack design that made the camera attractive to people. They kept buying it, perhaps out of inertia, even as that original simplicity increasingly disappeared. As soon, however, as a more “powerful” alternative design appeared, they were quick to recognize it and to recognize the shortcomings of the over-engineered Leica.

Attractive hypotheses also owe their seductiveness to their apparent simple explanatory power. Once that apparent simplicity is compromised, so, increasingly, is the hypothesis’s hold on the imagination. Eventually, the beauty of the idea as it was originally conceived is no longer to be found through the mass of overgrowth around it, the multiple qualifications and supplementary assumptions. When that happens, the rational enquirer no longer understands the “scientist’s” unwillingness to relinquish his original idea.

It is at this point that he asks the “scientist”: “Why are you holding on so desperately to your hypothesis? You must know that it’s barely recognizable anymore. It certainly no longer holds any intuitive plausibility”.

This point is politically important and interesting because the Left is characteristically marketing strongly counter-intuitive generalizations. Closer examination of these generalizations inevitably shows them to run afoul of contrary evidence. But when this contrary evidence is brought to the Leftoids’ attention, they introduce ad hoc “explanations” for as long as problematic data are produced; and, further, they do so with no regard to the probabilities associated with their ad hoc theories.

There are many examples of ad hoc-ing among the doctrines of the left, at least one current one is that of Global Warming, but you can find them wherever your instinctual response to a Left based generalization is: hmmm – that sure doesn’t sound true.

Your class assignment for the upcoming part of your life is to identify examples of generalizations presented as confirmed truth by people who will accept no amount of contrary evidence as sufficient to abandon them.

The use of the ad hoc hypothesis to retain a problematic hypothesis should be added now to 1) making one’s hypothesis theoretically unfalsifiable (compatible with any and all empirical findings), and 2) the use of the no-true-Briton move, in which what begins life as an empirical generalization is transformed into stipulative definition. Slowly, slowly catchee Leftie!