Let’s take a look at some of the difficulties I described before (What About Truth?) and see how a scientific approach could attempt to resolve them.   For example, even though we disagree with a Hanunoo about the color of freshly cut bamboo, we could settle the disagreement through Science.  We can measure the wavelengths of light, the spectrum, reflected by the bamboo, and this spectrum would be its color, whatever name we want to give it.  And we can repeat the measurement a few times to alleviate any concerns of a fluke.  Pretty straightforward, right?  Similarly, any confusion about a scorpion being a lobster can be cleared by taking a small piece of the animal, extracting and analyzing its DNA, comparing the results with those obtained from other animals.  Then we classify it according to what the result matches with, whatever we want to call it.  Again, pretty straightforward.

But when we actually go through the motions, questions begin to arise.  To measure the spectrum of reflected light we need someone who actually knows how to do it.  They will use a specialized piece of equipment, which we assume is working properly, that is, its performance has been checked to ensure it conforms to certain specified criteria.  Moreover, we assume that the person operating the equipment knows how to use it properly, that is, has been trained to use it and their ability to do so has been ascertained accordingly.  Similar considerations apply to using DNA analysis to decide whether an animal is a scorpion: we will need specialized equipment as well as someone who is trained in how to obtain the DNA and use the equipment to analyze it.  In general, in order to scientifically measure a feature of the world, be it the spectrum of reflected light, the kind of animal, or something else, we need tools and equipment along with a trained operator.  The operator has been taught the proper use of the equipment, the proper way of measuring the feature of the world we are interested in: they have been socialized by the scientific community that has developed the way of measuring and the equipment to use.

The knowledge generated by any scientific community has been developed through engagement with the material world, selecting and foregrounding certain features of experience while ignoring others, agreeing on which features to select as, say, the spectrum of reflected light, constructing equipment that would measure these features, and agreeing on what constitutes proper performance and operation of the equipment.  In the same vein, there is agreement on what constitutes DNA, how to operate equipment to analyze it, what patterns to accept as representative of a scorpion.  The organization of the experience of the world in terms of spectra or DNA rests on a constellation of abilities accepted and learned within particular scientific communities.  In other words, “this is a scorpion” or “this color is red” is always checked within a framework of abilities and socialization.

The point has been made before in very many ways, and is often referred to as the Duhem-Quine thesis.  Pierre Duhem pointed out that the truth of any statement about the world always rests on “auxiliary assumptions”, such as appropriate know-how, the proper functioning of equipment, and so on.  The result is that, if, say, we find that a statement fails to correspond to our experience of the world, we can always attribute the failure to some of the “auxiliary assumptions” and adjust these assumptions to rescue the truth of the original statement.  W.V.O. Quine made a similar argument (in “Two Dogmas of Empiricism”), pointing out that “our statements about the external world face the tribunal of sense experience not individually, but only as a corporate body,” and “Any statement can be held true come what may, if we make drastic enough adjustments elsewhere in the system.”

We can put this insight in more general terms, namely that truth and knowledge are embedded in and are part of the way in which we engage with the world.  And this insight encompasses all kinds of truth and knowledge, including the scientific.  A general way to express the basis for this insight is that in order to check the truth of any statement about the world, we need to know how to check it, we need to ensure that we check it correctly.  The methods and the abilities through which we check are established and accepted within the community in which we have been socialized.  This uncertainty about methods and abilities comes to the fore whenever we develop new knowledge: is our new finding a feature of the world, or we are doing something wrong?  Or, when we fail to come up with the new finds we were expecting, is it because they were not there or because we were not doing things right?