Philosophy Talk: The Blog: Does Truth Matter?

From the 'PhilosophyTalk' blog, Ken Taylor's posting on Does Truth Matter? leads to a discussion in which the question of whether the question has a well defined answer becomes one of the issues to address. But although "Truth matters" may not be a "complete proposition" in the sense that it is neither universally true or universally false, perhaps the emphasis of the original posting was more on the need (or not) for avoiding untruth, rather than on finding value in every true statement.

Sometimes the truth of a matter really does NOT matter or is best not known, but perhaps there is a stronger case for the thesis that "UNtruth matters". ie that it is almost always wrong to believe or assert that which is demonstrably false. (And perhaps the specific theory of truth being applied is less relevant if we restrict to the "demonstrable" situation where correspondence and convention coincide.)

So the question becomes under what circumstances may a falsehood not be a bad thing.
The finding of value in a "serviceable falsehood" of the kind exemplified by "Saddam had WMD", (as used to motivate soldiers into having a sense of purpose which may have enhanced their effectiveness), is however not related to any particular property of truth or falsehood. It is rather just another example of an "ends vs. means" issue, perhaps analogous to the argument of net utility that may be used by some in the familiar moral exercise of deciding whether or not to push someone off a bridge in order to block a train which would otherwise run over several people standing on the track. To some, doing wrong to produce an eventual good result can never be justified, but many others accept the net utility argument. For example many of us who would not push the fat guy off the bridge to block the train from running over the kids will regularly and willingly support minor injustices for some on the basis of serving the greater good (eg a not entirely equitable tax law for which the fairer alternative would be more expensive to administer). But even those would probably agree the small injustice is wrong "in and of itself", and should be avoided if the same general gain could be achieved in some other way. Similarly, the possible net value of a "serviceable falsehood" does not contradict the fact that promoting the falsehood is ("morally") wrong in and of itself. But anyhow, as I said at the beginiing of this paragraph, this aspect of the issue has nothing to do with the particular issue of Truth per se.

So my question is: Is there more to the idea of "serviceable falsehood" than this?

Scientific theories are often described as serviceable falsehoods which we accept for lack of a better alternative. This is what I believe Ken was getting at, although his reference to "approximate truth" may have led some of us astray. One view of a scientific theory is as something which claims only to compactly "predict" the results of all past observations (at least to within the accuracy range with which those observations were made). As such, if successful, the theory is true so long as its predictions all fall within the error bounds of the corresponding observed measurements. But when the theory is used to predict future observations, then it runs the risk of being falsified - as all theories will be (at least for so long as science continues to be worth doing). But "falsification" of a theory doesn't always make it false. Often, as in the case of Newtonian mechanics, it just puts restrictions on that theory's domain of validity.

An important distinction here (which appears to escape the compehension of many non-scientists) is between two entirely different notions of scientific 'theory'. One, like Newtonian mechanics or its various relativistic and quantum sequels, is a set of rules (generally expressed in mathematical formulae) relating various observed values, and the other is an explanation of some phenomenon in terms of a higher-level theory of the first kind. The statistical mechanical explanation of thermodynamics is an example of the latter, as is also just about every theory of astronomical or biological evolution.

It is in the latter case (of "theories" which purport to explain some observation) where, to a scientist, "truth" is actually at issue and "matters" (no matter whether anyone else actually gives a damn). There really is only one true answer to the question of how our solar system originated, and most scientists expect that eventually we will find sufficient evidence to confirm one such theory. The same applies to various questions about how certain steps in the evolution of current species occuurred, but although it would be foolhardy to suggest that there is no possibility of finding a purely mechanical path from non-life to the current situation, there are certainly cases where we do not yet know which "theory" is correct. And, yes, to those of us who care, it does matter a lot. (I don't know if Alexander Keith's Pale Ale is advertised in the US, but the tag-line is "Those who like it like it a lot" and perhaps the situation is similar.)

However a wrong theory of planetary or biological evolution is not a "serviceable falsehood". It is just plain wrong as a history of events, and it will eventually be found to conflict substantially with some observed fact to an extent not within the bounds of experimental error (and probably not even within any limited bounds that corresponded to the measurement capabilities of science at the time the theory was proposed).

Another kind of "serviceable falsehood" is promoted by some "enlightened" religious leaders. The thesis seems to be that the literal truth (or more probably untruth) of their scriptures does not matter because of some "deeper" meaning that belief in them is deemed to facilitate. This, I believe, is harmful, but the full extent of my reaction is best left unstated at this point.

Comments are closed.