Scientific Method in Practice (Pt. 5)

In this series of posts, I'm re-reading Hugh G. Gauch, Jr.'s philosophy of science textbook Scientific Method in Practice (Google Books).

[Series Index]

Recent History

Just as Part 2 of this series covered the four 'bold claims' of rationality, truth, objectivity, and realism, this post will focus on 'four recent woes' which 20th century philosophers have advanced: elusive truth, underdetermined theory, incommensurable paradigms, and redesigned goals.

Elusive Truth

I may be the only person whose first philosophy book was Karl Popper's Realism and the Aim of Science. His falsification criterion of separating science from pseudoscience appealed to me then, and I still frequently hear it invoked in discussions about science. Essentially, a theory or hypothesis isn't properly scientific if it can be used to explain just any data whatsoever; it must risk disconfirmation by implying what can't happen if it's true. This is what it means to be 'falsifiable.'

In Popper's day, the theories of Marx and Freud 'made sense' to huge numbers of people but seemed capable of interpreting anything as a validation. Meanwhile, Einstein's theory of relativity was very counter-intuitive, yet it made risky predictions that kept failing to fail! This limit (or 'demarcation') on proper science has been taken to heart in contemporary debates on whether evolution, intelligent design, climate change, etc. are scientific views.¹

But wait, where does the 'elusive truth' woe come in? From overcompensating to the point of denying any place for induction in scientific method. Notice how falsifying a hypothesis is a deductive affair: if P then not Q; Q; therefore not P. In Popper's view, science can only proceed in this negative fashion of eliminating false theories, without any way to add positive weight to the likelihood a theory is true.

Scientific theories can never be 'justified', or verified. But in spite of this, a hypothesis A can under certain circumstances achieve more than a hypothesis B—perhaps because B is contradicted by certain results of observations, and therefore 'falsified' by them, whereas A is not falsified; or perhaps because a greater number of predictions can be derived with the help of A than with the help of B. The best we can say of a hypothesis is that up to now it has been able to show its worth, and that it has been more successful than other hypotheses although, in principle, it can never be justified, verified, or even shown to be probable. This appraisal of the hypothesis relies solely upon deductive consequences (predictions) which may be drawn from the hypothesis. There is no need even to mention induction.²

As a toy example, let's say I have a small bag of marbles. After shaking the bag, drawing out one marble, seeing it's red, and putting it back in the bag, I conjecture that all the marbles are red. If the next marble I draw out is blue, my conjecture is refuted. But what if I draw out a red marble the next 10 times? The next 1,000 times? The next 1,000,000 times?

From a purely deductive point of view, we are no more justified in thinking the bag contains all red marbles after a million trials than after one. The conjecture that the bag is a half-and-half mix of red and blue marbles is equally justified. Apply this to scientific method in general and we must give up our confidence that well-tested theories are close to the truth.

Underdetermined Theory

Two points on an XY coordinate graph. Which equation fits the points? If you answered 'infinitely many equations can fit those points,' you win! This holds true as the number of points increases. As a computer graphics textbook puts it:

Given n data points that are numbered P₁ through P_n, there are infinitely many curves that pass through all the points in order of their numbers, but the eye often tends to trace one imaginary smooth curve through the points, especially if the points are arranged in a familiar pattern. It is therefore useful to have an algorithm that does the same.³

I hope the analogy to observations (points) and scientific theories (equations) is clear. As human beings, we may tend to prefer a certain style of theory 'fit' to our observations, but there are an infinite number of other possible theories which can fit the data. In other words, the data does not uniquely pick out (determine) a theory. We'll come back to this idea in a moment.

In his book Patterns of Discovery: An Inquiry into the Conceptual Foundations of Science, Norwood Russell Hanson argued that observation is dependent on theory (i.e. observation is theory-laden).⁴ This threatens to systematically undermine the distinction between theory and observation. And if that happens, then the problem of underdetermined theory produces the problem of underdetermined observation.

What does this mean for falsification? It is reduced to a mere coherency check between the theories underlying an observation and the conjectured theory. A contrary observation cannot prove a theory false, only inconsistent with some other theories.

So the first woe cited earlier was that science could not verify truths. Now this second woe is that science cannot falsify errors either. Science cannot declare any theory either true or false!⁵

Incommensurable Paradigms

In The Structure of Scientific Revolutions, Thomas S. Kuhn outlined a new model for the way science changes over time. Instead of seeing science as steady progress occasionally hastened or corrected by geniuses, he identified two radically different modes.

During periods of normal science, the scientific community tacitly agrees to a set of assumptions and works on a set of puzzles they currently find interesting. They make steady progress on these puzzles using their shared assumptions. This combination of the (current) way science is done and the (current) objects of scientific inquiry constitute a paradigm.

Periods of revolutionary science begin when new assumptions or new puzzles disrupt the old fit between the two. There is no rational way to say one set of assumptions is more correct than another, or that one set of puzzles is more important than other. When science finishes shifting to a new paradigm, it's a matter of changing to match circumstances not improving against a common measure.

The paradigms of two periods of normal science are incommensurable, which means not-commonly-measurable. So why do we tend to think Einstein's physics was an improvement over Newton's physics, and Newton's physics an improvement over Aristotle's physics?

Revolutions close with a total victory for one of the two opposing camps. Will that group ever say that the result of its victory has been something less than progress? That would be rather like admitting that they had been wrong and their opponents right. To them, at least, the outcome of revolution must be progress, and they are in an excellent position to make certain that future members of their community will see past history in the same way.⁶

To adapt the old saying: the victors write the science books.

Redesigned Goals

Gauch's name for this 'woe' does not fit well with how he describes it. Reassigned Role would be better.

This final woe is the natural result of all of the above. If science can't discover truth, or eliminate false theories, and only reflects the current preoccupations of a cloistered social group, then the high role it plays in modern thought is unjustified. Science is an ideological tyrant oppressing other ideas, and it's high time for us to drag science down from its stolen throne.

Paul Feyerabend's essay 'How to Defend Society Against Science' is very much worth reading, especially knowing the historical background I've given above. These same arguments pop up constantly in popular discussions when someone wants to shield an idea from scientific criticism (see Tim Minchin's poem Storm). I'll close out this post with Feyerabend's view on how government should treat science and remind you that this is a perfectly reasonable response if the 'woes' in this post are accepted:

Science is just one of the many ideologies that propel society and it should be treated as such (this statement applies even to the most progressive and most dialectical sections of science). What consequences can we draw from this result?

The most important consequence is that there must be a formal separation between state and science just as there is now a formal separation between state and church. Science may influence society but only to the extent to which any political or other pressure group is permitted to influence society.⁷

1. An example of falsification's entrenched role in popular debate: http://www.conservapedia.com/Falsifiability_of_evolution
2. Popper, K. (1935/2005). The logic of scientific discovery (3rd edition). Routledge. p. 317. (italics original; bolding added)
3. Salomon, D. (2006). Curves and surfaces for computer graphics. Springer. p. 141.
4. Hanson's own introduction of this idea is very readable: http://evans-experientialism.freewebspace.com/hanson.htm
5. Gauch, H. G., Jr. (2006). Scientific method in practice. Cambridge: Cambridge University Press. p. 84
6. Kuhn. T. S. (1962/1996). The structure of scientific revolutions. Chicago; The University of Chicago Press. p. 166.
7. http://uwacadweb.uwyo.edu/Ashleywy/paul_feyerabend.htm

On 'Metaethical Contextualism Defended' (Pt. 2)

(...continued from Pt. 1)

To review, contextualism about the meaning of 'ought' is the view that...

1. The word 'ought' can mean different things, i.e. make different contributions to the proposition expressed by a sentence.
2. 'Ought's meaning is completed by some features of the context in which it's used.

We can see similarities in the use of pronouns. The word 'I' in the abstract doesn't refer to anyone, but when I say 'I' it refers to me! (And when you say 'I' it refers to you.) The context of its use completes its meaning.

According to Björnsson and Finlay (B&F), the meaning of 'ought' in the abstract is incomplete in two ways:

We believe that normative “ought” claims are doubly relative to context, being relativized both to (i) bodies of information and (ii) standards or ends. On this view, every meaningful normative utterance of a sentence “A ought to φ” will express a proposition to the effect that A ought-relative-to-information-i-and-standard-s to φ, for some i and s determined by the context of utterance.¹

My last post was about the information component; this time we'll look at standards contextualism.

General Normative Oughts

Charles, you ought to head east for five miles, then take the first right after the bridge.

Is the above statement true? It depends on the goal implied by context. If this advice is about Charles getting to the nearest gas station — and this is the best way — then it's true. It's like telling a student she ought to read page 15 of the assignment extra carefully; there's a goal (or end) involved with that 'ought.'

Contrast this with the idea of 'just plain ought' (aka ought simpliciter). If we say Charles just plain ought to drive in such-and-so a direction, and the student just plain ought to read a certain page carefully, what are we even claiming? I've never been able to make sense of the notion that a person ought to do something while explicitly refusing to qualify that 'ought.' At best, we might be able to claim a person ought to do something in order to act rationally, or morally, or legally. But these are all standards (or kinds of standards) and this fits neatly into the contextualist view that meaningful 'oughts' are relative to standards, ends, goals, and the like.

The take-home lesson is that we often use qualified 'oughts' without spelling it out. What we mean can usually be discerned from context.

Single and Multi-Standard Morality

Let's suppose there is only one moral standard. When we use 'ought' in various non-moral senses, different standards may be understood from context; but moral 'oughts' are always relative to the same standard. This situation still favors semantic contextualism over invariantism (because the meaning of 'ought' varies by context), but it wouldn't be metaethical contextualism.

Björnsson and Finlay take things one step farther. They believe moral 'oughts' can also vary in meaning by implying different standards. If so, this would explain "why moral beliefs diverge between cultures, and why moral disagreement persists among well-informed competent judges."

Simply put: people can mean different things by superficially identical moral 'ought' claims because they're going by different standards.

Challenge #2 — Moral Disagreement

The second half of B&F's paper defends this view against the objection that it doesn't account for moral disagreement.

Take the example of Huckleberry Finn thinking he ought to tell authorities about the runaway slave, Jim.² Modern Americans would almost all insist it's not true that Huck ought to tell on Jim. If there were a single moral standard, then at least one of these claims must be false. However, if Huck and modern Americans are using moral 'oughts' which derive their meanings from different standards, then both claims could be true!

According to a contextualist treatment of standard relativity in normative judgment, it is possible that the propositions that Huck accepts and that we reject are different: Huck accepts the proposition that he ought-relative-to-standard-Y to tell on Jim, while we reject the proposition that he ought-relative-to-standard-Z to tell on Jim. So it seems that contrary to appearances we are not really in disagreement with Huck: what he accepts is not what we reject.

But it would be hard to take a moral theory seriously that denies we have moral disagreements. We certainly think we have them!

Two Ways to Disagree

When arguing a moral point, it can be helpful to argue from the other side's values. David Boonin takes this approach in his book A Defense of Abortion, i.e. he argues from the values of pro-lifers that abortion is morally permissible (or at least that the values of pro-lifers don't entail abortion being morally forbidden). To use the above terminology, pro-lifers say one ought not abort relative-to-standard-Z and Boonin disagrees by claiming it's false that one ought not abort relative-to-standard-Z. This is a logical conflict because one side is asserting a proposition (P) and the other is denying the same proposition (not P).

Not all disagreements need to be logical conflicts. We can also have a conflict in attitude that arises from holding different values. We can agree Huck would be acting consistently with the prevailing moral standard of his society if he told on Jim, but still want him not to tell on Jim because of our own valuing of human freedom.

I would characterize moral claims as typically two-pronged: expressing an attitude toward a standard and expressing a belief about an action's relationship to that standard. Moral disagreements can be about either or both of these aspects.

Truth Assessments

[C]ontextualism would seem to bar us from expressing our disagreement with Huck by saying that his belief is false, and force us to say that it is true.

B&F respond to this by granting the truth of "Huck believed that he ought to tell on Jim" while denying "Huck knew that he ought to tell on Jim." They draw a parallel to the story about information contextualism to show that a person in a position to know better can affirm the fact of another person's belief without counting it as knowledge. It would be like the police officer agreeing that the rescue commander believes the child might have gone 'downstream,' but denying that the commander knows the child might have gone 'downstream.'

I have trouble with this analogy because we're not in a position to know Huck is making a true/false kind of mistake (we're actually operating under the assumption that he isn't). Instead, I would grant that Huck both believes and knows that he ought to tell on Jim — where 'ought' is relativized to Huck's standard — but we're still justified in denying that "Huck ought to tell on Jim" because now 'ought' is being relativized to our standard.

In other words, Huck's belief is a filled-in proposition that we must recognize as true. Yet the sentence "Huck ought to tell on Jim" does not necessarily express Huck's belief (thanks to contextualism), so we aren't forced to say it's true.

Concluding Remarks

The main purpose of 'Metaethical Contextualism Defended' was to address a few objections to the view, not so much to explain the positive motivation for accepting it. Nevertheless, B&F take time to highlight three major reasons to prefer contextualism over invariantism:

• 'Ought' does sometimes seem to require a standard/end/goal/etc. (See my section on General Normative Oughts above.) It's simpler to assume all 'oughts' function in this way.
• When anthropologists and others are merely reporting on morals, a relative-to-standard-s treatment can become flagrant. So, again, it's simpler to think this is always going on even when downplayed or denied by those making judgments according to the standards they themselves support.
• Contextualism explains what moral claims mean in a non-mysterious way (promoting standards and judging how actions relate to them), why moral disagreements are such a pain (two types of disagreement in one!), and why moral judgments are typically motivating to those who make them (they're usually relativized to standards one personally supports).

As I mentioned up front, I'm interested because I'm already on board with metaethical contextualism. Writing these posts helped me work through a few of the objections, though I'm sure other criticisms will be put forward as awareness of this view grows.


1. Björnsson, G., Finlay, S. (2010). Metaethical contextualism defended. Ethics 121. pp. 7-36.
2. I'm using the paper's example this time.

On 'Metaethical Contextualism Defended' (Pt. 1)

In this paper, Gunnar Björnsson and Stephen Finlay (B&F) defend a form of metaethical relativism called contextualism against arguments that it fails to make sense of common practices like giving advice and expressing moral disagreement. This may seem rather niche, but I'm interested because I think contextualism is the best explanation of moral language.

Metaethical Contextualism Defended — Gunnar Björnsson and Stephen Finlay (search link)

The focus here is on the word 'ought.' Three views of 'ought' sentences are contrasted:

Invariantism. The 'ought' in ought sentences always contributes the same meaning to the proposition it expresses. Propositions have objective truth value.

Relativism. Like invariantism as far as 'ought's contribution to the proposition is concerned, but propositions themselves have subjective truth values.

Contextualism. The 'ought' in ought sentences is 'semantically incomplete' and so its meaning can vary. This in turn can produce different propositions with correspondingly different objective truth values.

Note: The above relativism is semantic relativism. Both contextualism and semantic relativism can be used in metaethical relativism. I think 'metaethical contextualism' is just a shortcut for saying 'metaethical relativism with contextualist semantics.'

Challenge #1 — Deliberation and Advice

The contextualist theory B&F support says that normative 'ought' claims are "doubly relative to context, being relativized both to (i) bodies of information and (ii) standards or ends." In this post, we're going to look at information relativity, the idea that 'ought' can take on different senses depending on which body of information is salient in the context of the claim.

A Rescue Story

Suppose there is a report of a child entering a storm drain and heavy rainfall is on the way.¹ The rescue commander can choose to send her full search team 'upstream' from where the child entered, 'downstream,' or split the team in half. Based on the information she has, she estimates the chances of locating the child before the team has to evacuate as follows:

Upstream — 80% if child went that way; 40% overall
Downstream — 100% if child went that way; 50% overall
Split Up — 60% whichever way the child went

The commander concludes that since 60% is the highest overall chance, she ought to split the team. Seems like the correct judgment, right?

Agent's Information vs. Full Facts

In fact, the child went 'upstream.' If the commander splits the team, the child has a 60% chance of being rescued, but if the full team were sent 'upstream,' the chance would rise to 80%. It seems wrong now to say the commander ought to split the team. Instead, she ought to send the whole team 'upstream.'

What caused the shift in what the commander ought to do? Let's look back at our three semantic options:

Under an invariantist interpretation, both 'ought' judgments can't be true; the commander's judgment is just plain incorrect. Even though splitting the team seems the best strategy, given her information, the only true 'ought' is that she ought to send the whole team 'upstream'...the worst strategy from what she can tell!

Under a relativist interpretation, the judgment that she ought to split the team and the judgment that she ought not split the team are in direct contradiction, but both are true from different standpoints.

Under a contextualist interpretation, these two judgments are not in direct contradiction. Instead, the 'ought's take on different senses. Given only the commander's information, it's true that she ought to split the team; given the additional fact of which direction the child went, it's true that the commander ought not split the team.

I hope you agree the contextualist interpretation comes off looking pretty good. However, there is a significant puzzle for contextualism here.

Correcting the Commander

Before the search team enters the drain system, a police officer arrives and interrupts: "I heard you want to split up the rescue team. Don't do that! Instead, you ought to send the whole team 'upstream' from here, because I just came from that direction and I saw the child down through a small grate."

Sounds like a natural enough thing to say. But remember, according to B&F's contextualism the police officer's 'ought' is in a different sense than the 'ought' used by the commander when she decided she ought to split the team.

So how does the new advising 'ought' even address (let alone disagree with) the original 'ought'?

Maybe the relativist interpretation is correct after all, since the two of them would at least be talking about the same sense of ought! ...or so the argument goes.

Contextualism Defended

Let's look at two ways the commander could deliberate:

1. She could ask, "What ought I do, given the information I now possess?"
2. She could ask, "What ought I do, given the information we all possess?"

The second option is a more accurate picture of what goes on in a shared deliberation, but it still doesn't help with the police officer's unsolicited advice. He showed up after the commander concluded she ought to split up the team. This is the "problem of advice from unexpected sources."

Here's the dilemma:

If the police officer was included in the original group referred to in "the information we all possess," then the commander was wrong to ever think she ought to split the team...even if she never heard from the police officer.

If the police officer was not included in this group, then we're back to 'ought' having different senses with "apparently no common question with which both are concerned, and deliberation and advice come apart in a puzzling way."

One solution is to add another option to the list above:

3. She could ask, "What ought I do, given the information I have now, plus what I gather while there is still have time to decide?"

This is news-sensitive contextualism. Instead of indicating a fixed group of other people who can contribute information, the agent is really interested in whatever information she can get from expected and unexpected sources. Why? Because she's not primarily motivated by the abstract question of what she ought to do relative to what some fixed group of people believe, but what she can do to have the best chance of rescuing the child.²

An Alternative Defense

My own response to this puzzle would be to drop B&F's 'bodies of information' contextualism while retaining the 'standards or ends' contextualism. The sense of correctness about the commander's original judgment that she ought to split the team can be understood as her drawing the most justified belief from her information. However, it was always true that she ought to send the whole team 'upstream.'

Why not let general epistemology take care of all the concerns about limited information, deliberation, and advice? Whether we deliberate internally, or in a group, or stay open to unexpected sources, it seems we're always trying to find out what we ought to do given the facts.

EDIT: In the comments below, Stephen Finlay points out a critical problem with this answer. I now accept contextualism rather than invariantism about information, for the word 'ought' anyway.

In the next post, we'll look at contextualism about standards rather than contextualism about information.

(...continued here...)


1. Story adapted from the somewhat more complex story about miners in the paper. (Ethics vol. 121.)
2. I'm mixing B&F's answers to both horns of the dilemma.