Facts and Values

by Kieran Healy on November 2, 2009

I recall a short but striking conversation with the formidable Piero Sraffa at the Economics Faculty cocktail party after Dennis Robertson’s Marshall Lectures. I well knew that it was Sraffa whom Wittgenstein had described as his mentor during the gestation of the Philosophical Investigations, but I still ventured a rather simple-minded remark about the obvious importance of the fact-value distinction to the social sciences. He turned on me his charming smile and glittering eyes. Did I really suppose that one could switch from fact to value as if simply moving a handle? His voice rose and his Italian accent grew sharper. “Fact, value! Value, fact! Fact, value! Value, Fact! FACT, VALUE! VALUE, FACT!” I beat a swift and chastened retreat. — W.G. Runciman, Confessions of a Reluctant Theorist, 18.



Jimbo 11.02.09 at 4:36 pm

Sraffa was da bomb. “The Production of Commodities by means of Commodities” is a thin little volume that utterly demolishes 95% of what is taught in the average economics department…


Hidari 11.02.09 at 4:43 pm

Sraffa was probably much more of an influence on the later Wittgenstein than isn generally thought, as is ‘Marx-as-interpreted-by-Sraffa’ although since the influence was mainly via conversation, academic scholars have generally overlooked it (also because philosophers in the Anglo-American tradition don’t like the idea that ‘our’ Wittgenstein was influenced by that beastly Marxist nonsense).

Incidentally, the fact-value ‘distinction’ is clearly and self-evidently bollocks.


dsquared 11.02.09 at 4:51 pm

Incidentally, the fact-value ‘distinction’ is clearly and self-evidently bollocks

is that a fact?


geo 11.02.09 at 5:24 pm

is that a fact?

Two-thirds fact, one-third value.


Richard J 11.02.09 at 5:28 pm

And a few drops of angoustora bitters.


P O'Neill 11.02.09 at 6:05 pm

I want back the many hours spent trying, unsuccessfully, to understand “The Cambridge Controversy”.


kid bitzer 11.02.09 at 7:11 pm

His voice rose and his Italian accent grew sharper. “Fact, value! Value, fact! Fact, value! Value, Fact! FACT, VALUE! VALUE, FACT!”

damn. why din’t i ever think of that argument?


Will Davies 11.02.09 at 7:35 pm

Wittgenstein wrote somewhere (and, to be even less helpful, I paraphrase) that if ever there were a book of ethical philosophy that was actually about ethics, it would explode every other book ever previously written.


Carl 11.02.09 at 8:06 pm

Sraffa was also a good friend and mentor to Antonio Gramsci; that’s an interesting little triangle right there. Perhaps one of those people whose genius is not so much genesis as catalysis.

But I think the ‘switch’ argument is a bit of a red herring, which may well be Runciman’s subtext. At least the way Weber deployed it, the fact/value distinction was not meant to be a description of an objective state of affairs or even a real possibility, but an ideal type of an ethical relationship to knowledge. The ‘vocation of the scientist’, something like ‘getting it right’. So the idea is that if we can see how values are normally mixed with facts (after first defining our terms, ‘fact’, ‘value’) and correct for that, it might be possible to construct knowledge that achieves a certain reliability and respectability beyond, or other than, the particular identity and/or power of its author(s).

In a wittgensteinian sense the fact/value distinction is accordingly a move in the game ‘science’, and as such a criterion for the production of a certain kind of ‘truth’. So Runciman was quite right, but in a narrow way.


Billikin 11.02.09 at 8:24 pm

“Truth is beauty, beauty truth.”


Doctor Science 11.02.09 at 8:28 pm

Totally OT for this discussion, but not for this blog:

PhD Comics creator almost deported from the UK.


ejh 11.02.09 at 9:05 pm


Salient 11.02.09 at 10:11 pm

I value facts. And that’s a fact. (Which I value.)


Salient 11.02.09 at 10:12 pm

Well, shucks, a bit more seriously:

Wittgenstein wrote somewhere (and, to be even less helpful, I paraphrase) that if ever there were a book of ethical philosophy that was actually about ethics, it would explode every other book ever previously written.

…Are you completely positive that wasn’t Heidegger?


Phillip Hallam-Baker 11.02.09 at 11:52 pm

One of the things that some people have great difficulty with is the fact that there are limits to what can be deduced using a system of logic. They simply cannot process the fact that mathematics does not answer every question of importance.

And so there was that student of Robert Nozick who was trying to establish a logical positivist foundation for Zen Budhism at one time. Now how is that for a koan – beats one hand clapping eh?

Now if we just accept as fact that the deductive power of logic is not going to let us deduce a complete system of ethical values from Peano’s axioms, that we are going to have to bias the system with at least some starting assumptions, we can actually get some useful work done.

If we agree that the starting assumptions might be something we have to agree on through some form of consensus forming process then we may get somewhat further still.

The reason people seem to resist this approach is that it does not offer a result that can be used to beat people round the head who roundly deserve same for pontificating systems of value that range from obnoxious to downright evil. Contrawise, said pontificants despise the approach because it does not give them ammunition.

Seems to me that the real value of this type of debate is that it completely shuts the door on ethical systems whose foundations are ‘theological’. Here I will lump together all arguments that go from a ‘fact’ (e.g. the sun rises each morning) to a ‘value’ (e.g. therefore we have to cut the hearts out of people each night to make sure this continues). While the Mayan priesthood is out of business these days, there are plenty of priests left making similarly spurious arguments.


Salient 11.03.09 at 12:22 am

Phillip Hallam-Baker!

Whatever happened to Quantum of Stupid? [I still tell folks that a power drill and a copy of Windows XP is required for Ubuntu installations.]


Phillip Hallam-Baker 11.03.09 at 12:40 am


Well, err, I took the studio apart for my other video project ‘Phill’s British Workshop’. It is kind of modeled on the New Yankee Workshop, only with slightly more tools and a lot more daleks.

Should be starting up again, maybe.

Now the Ubuntu folk are complaining that the problem I was having was due to my buying the cheapo 650Mb CDR media rather than the 700. Which they might have a point on, only what my real complaint was there was the fact that 90% of sysadmin time is spent dealing with unnecessary problems that are really simple after the fact.

Like who on earth would care about the capacity of CDR media or have noticed that they updated the spec? It is an utterly obsolete format as far as I am concerned. The only time I ever burn a CD or DVD is to install something.

Actually, the QOS I do need to do soon is on the infamous MacBook Air hinge defect which turns out to be a complete design error. Its a doozy of a design error. Basically the MacBook Air will self destruct if you use it on an airplane. If you look at the design of the hinge, there is a cam that stops it opening too far. But when you have the thing fully open (easy on an Air, they don’t open very far) the cam turns into a pivot about which the weight of the display is acting on the hinge with a 1:100 leverage ratio. That means you have 100 times the weight of the display pulling the cheap pot metal hinge apart.

So the weight of the screen vibrating against the cam will basically destroy the hinge after a couple of airplane trips.

Is that a ‘fact’ or a ‘value’? Hey, I am even further off base this time than the first try!


vivian 11.03.09 at 1:00 am

Windows might be a fact, Ubuntu might be a fact, but I’m thinking that Macs are values. (May Hilary Putnam forgive me for such things)


Phillip Hallam-Baker 11.03.09 at 1:03 am

Nah, Macs cost way too much to be considered in the value category.


jsm 11.03.09 at 1:16 am

Perhaps we would learn something from going down to some kind of fundamental vacuum level where there really are neither facts nor values:


Or perhaps not.


Rosa Lichtenstein 11.03.09 at 1:36 am


“also because philosophers in the Anglo-American tradition don’t like the idea that ‘our’ Wittgenstein was influenced by that beastly Marxist nonsense”

In fact, analytic philosophers have long been aware of this influence — how could they fail to be if Wittgernstein himself indicated this in the ‘Investigations’?

I have outlined these influences here:


See also:

Kitching, G., and Pleasants, N. (eds.), ‘Marx And Wittgenstein. Knowledge, Morality And Politics’ (Routledge, 2002).


bad Jim 11.03.09 at 5:09 am

It is said of Niklaus Wirth that Europeans tend to pronounce his name properly, as Nih-klaus Virt, while Americans usually mangle it into something like Nickles Worth, which is to say that Europeans call him by name while Americans call him by value. (This refers to an issue in the design of programming languages.)

Russell was unimpressed by a family member who dismissed philosophy with “What is mind? No matter. What is matter? Never mind.”


Fr. 11.03.09 at 12:45 pm


I am planning to use that during my viva to answer any question on data integrity.


Duncan 11.03.09 at 2:28 pm

Wittgenstein on the exploding ethics book and fact/value distinction can be found here:



Patrick S. O'Donnell 11.03.09 at 2:46 pm


Phillip Hallam-Baker 11.04.09 at 12:13 am

I know of at least one Turing award winner who has been known to pronounce Wirth’s name as ‘Worthless’.

Trying to write string handling routines in Pascal was impossible.


Tim Wilkinson 11.04.09 at 1:39 am

Fact is value, value fact.


Phillip Hallam-Baker 11.04.09 at 4:34 am

That is the ACM Turing Award, not a ‘Turing test’

Leave it to comp sci to conflate the name of their most prestigious award with a CAPTCHA thingie.

But returning to the original discussion, while facts and values might exist in some Platonic universe, all we can deal in are assertions. And while in the Platonic universe there may be ‘factlike’ and ‘valuelike’ assertions, the assertions we have require interpretation through a lens that is clouded by factlike and valuelike biases.

Perhaps Discovery Channel or TLC could do a reality TV series on the cave.


JoB 11.04.09 at 9:24 am

Which goes to show (call it the Law of Yodaïsm) that any statement centering on just 2 nouns will tend to be funny, or deep (the quantum state is decided by the context).

I always thought I should read Sraffa; I’ll let that self-made obligatiob stand but count myself lucky never ever to have to talk to the prick.

(one proof point for the Law of Yodaïsm: the continueing vogue of haiku’s)


alex 11.04.09 at 9:30 am

“a CAPTCHA thingie”? You don’t know much about the history of computing, do you? Or maybe you do, and this is some bizarre involuted sarcasm I can’t quite grasp?


dsquared 11.04.09 at 9:46 am

#29: I haven’t checked the IP addresses (and indeed don’t know how to and if Kieran is remotely sensible, don’t have the requisite editor privileges either), but the Phillip Hallam Baker who is chief scientist at VeriSign does comment here occasionally, and so it is quite possible that he is joking.


alex 11.04.09 at 10:29 am

Oh, good. I suppose that means we are free, for the moment, from confrontation with the winners of Turing tests, as opposed to Turing awards…


stostosto 11.04.09 at 11:50 am

You can value a fact, but you can’t fact a value. And that’s a proposition.


stostosto 11.04.09 at 11:52 am

Also, I haven’t got a clue about what this is all about. And that’s an observation.


Rosa Lichtenstein 11.04.09 at 12:57 pm

So, what was so objectionable about my post that it was rejected?

Only a guess, but your pseudonym is (or was) notorious as a tax haven, and might therefore be a likely spam word – John Q


Walt 11.04.09 at 2:00 pm

stostosto: Nobody wants to spell it out explicitly, but alex and dsquared have discovered that the chief scientist at VeriSign is a robot.


Rosa Lichtenstein 11.05.09 at 7:03 pm

Ok, fair enough.


Phillip Hallam-Baker 11.06.09 at 1:50 pm

After an earlier post in the thread, I had someone concerned that there might be confusion between Turing Award and Turing Test. A CAPTCHA is probably the best know use of an attempt to create a Turing test.

I no longer work for the company. I don’t think they replaced me with a robot, but it is true that I am building daleks in the basement, so that is probably where the confusion arises.

Comments on this entry are closed.