Lesson 6 of 7
The empirical objection, in full
If no observation can overturn a theory's core, it is a different kind of thing from one that can. The objection, at full strength.
By Javier Sanz
Founder of ValueScreener. Built Ninety Nine, a securities services provider, and sold it to a tier-one global bank; ran operations and internationalization at Alpaca, the brokerage API.
Who writes thisEditorial processLinkedIn (opens in a new tab)X (opens in a new tab)
In one minute
- The objection is not that the deductions are invalid. It is that a body of claims no observation can touch is not doing the job a scientific theory is usually asked to do.
- A second objection bites harder: several propositions presented as deduced turn out to carry hidden assumptions about the world, which are exactly the kind of thing evidence can reach.
- Neither objection touches every part of the tradition. Hayek's work on prices and knowledge is empirical in temper and is not defended on these grounds.
The previous lesson stated Mises's claim that the core of economics is deduced rather than measured. This lesson states the objection, and states it at its strongest rather than at its most convenient.
What a claim gives up when nothing can count against it
A claim evidence can reach
- States in advance what would count against it.
- Can be wrong, and can be shown to be wrong.
- Disagreements can in principle be settled.
- Accumulates corrections over time.
A claim evidence cannot reach
- Nothing observed is inconsistent with it.
- Cannot be shown to be wrong by any measurement.
- Disagreements are settled by argument alone.
- Errors survive, because nothing removes them.
The complaint is about what kind of claim it is, not about whether the deduction is valid.
The first objection: immunity is not strength
If a proposition is true before any data is collected, no data can correct it. That sounds like a strength and the objection is that it is the opposite. A body of claims that no observation can touch cannot be improved by observation either, and the ordinary process by which a discipline notices it was wrong stops working.
There is a sharper version. Mises's argument is that his propositions follow from the fact of action. If that is right, then they are true of any world at all, including worlds nothing like this one, and a claim true of every possible world says nothing in particular about ours. The more securely the deduction is made, on this reading, the less it tells you.
The second objection: the assumptions are smuggled
This one is more damaging because it does not need any position on philosophy of science. Caplan, in the Southern Economic Journal in 1999, examined several claims the Mises and Rothbard tradition presents as following from the nature of action, and argued they do not follow at all.
His clearest case is indifference. Mises and Rothbard hold that a person can never be genuinely indifferent between two options, because indifference could not be shown in a choice. Caplan's reply is that the argument needs an extra premise nobody defended: that every preference a person has must be capable of being revealed in an action. Drop it and indifference is unremarkable. A person can know by introspection that they do not mind which of two shirts they take, pick one, and be indifferent all the same. Caplan notes that the tradition elsewhere rejects exactly this kind of behaviourism, so it is not available here either.
He makes parallel arguments about continuity and about how the tradition treats imperfect knowledge, and concludes that the attempt to rebuild economics on foundations of its own does not succeed.
What the objections do not reach
They are aimed at the Misesian core. They do not land the same way on Hayek, who argued in a different temper: his 1945 paper on knowledge is a claim about how an actual economy uses information that people do not hold in one place, and it is defended by argument about observable arrangements rather than by appeal to prior certainty. That difference inside the tradition is real and it matters for the rest of this track.
Check yourself
What is the first objection to a body of claims that no observation can touch?
- The deductions are logically invalid
- Nothing can correct it, so the discipline loses its way of noticing error
- It cannot be written down precisely
Show the answer
Answer: B.
The objection accepts the deductions and questions what kind of claim results. Immunity from evidence also means immunity from correction.What does Caplan argue the rejection of indifference actually depends on?
- A claim about the mathematics of utility functions
- An extra premise that every preference must be capable of being revealed in action
- A measurement of consumer behaviour
Show the answer
Answer: B.
His 1999 paper argues the conclusion needs that premise and that the premise is undefended, and that the tradition rejects the behaviourism it would require.Which part of the tradition do these objections reach least?
- Mises on the a priori character of the theory
- Rothbard on preference orderings without ties
- Hayek on prices and dispersed knowledge
Show the answer
Answer: C.
Hayek's 1945 argument is about how an actual economy uses information nobody holds centrally. It is defended in a different way and is not immune to evidence in the same manner.Sources
- Caplan, 'The Austrian Search for Realistic Foundations', Southern Economic Journal, 1999, on indifference, on preferences not revealed in action, and on the wider claim that the alternative foundations fail.
- Mises, Human Action, 1949, chapter II, for the position being argued against.
- Hayek, 'The Use of Knowledge in Society', American Economic Review, 1945, as the contrasting style of argument inside the same tradition.
What this leaves out
- This is a heterodox tradition. The mainstream account of the same phenomenon differs, and nothing here is a forecast or a view on any security.
- This lesson argues against the method of the previous lesson. That is deliberate, and neither lesson is the verdict of this track.
- Replies to Caplan were published by economists inside the tradition. They are not summarised here, and a reader who wants the dispute whole should read both sides.