Is Henry George's Political Economy Scientific?

Harry Pollard

[Reprinted from a Land-Theory online discussion, March 1999]

Cliff Cobb wrote:

The recent exchange with Harry about his definitions and concepts and the conclusions he derives from them, without any reference to what Karl Popper called "falsifiable propositions," has reminded me of one of my pet peeves: the widespread belief among Georgists that HG's political economy is "scientific."

Response from Harry Pollard:

One doesn't draw conclusions from defined concepts. One may however, use the developed language as a tool with which to draw conclusions. Why should this process require me to refer to Karl Popper and his "falsifiable propositions"?

Are you suggesting that a science should proceed from its beginning with no attempt to affix terms to phenomena. You are anxious to introduce mathematics to a field demonstrably ill-fitted for it. When Marshall, et al, did this at the turn of the century, they failed -- and their successors have been failing ever since.

Essentially, you are saying that you prefer mathematical language to our normal communicative language. That's fine -- but before you pay with your As and Bs and Xs would it not be a good idea to explain what you mean by them?

If you feel obliged to do so, I would advise you to have them stand in for well defined mutually exclusive concepts. Or, would that not be 'scientific?

I have laid out the beginnings of the Science of Political Economy. Your contributions up to now seem to be like this comment (it isn't really a contribution):

Cliff Cobb wrote:

Not only is that assertion of questionable factual accuracy, it also tends to support an anti-empirical bias that I have observed on this list. ("Numbers? We don't need no steenking numbers.")

Response from Harry Pollard:

You've said this kind of thing before. So, what is your empirical need? You told us.

Cliff Cobb wrote:

"All I want are rough approximations."

Response from Harry Pollard:

As this piece was in answer to my argument that we cannot measure Rent in the present economy apparently, after casting doubt on the accuracy of my statement, and making a little snide at the list -- you agree with me that accuracy isn't possible.

However, I was as usual, kind, saying "who could argue against a rough guess?"

Once you had cast doubt on the scientific propriety of establishing a language, you proceed to tell us the language you use.

Cliff Cobb wrote:

"Those are not arbitrary or private definitions. I am following conventions, the same ones that Victor appears to be following."

Response from Harry Pollard:

It was the inadequacy of such definitions that led George carefully to define the language he was using. Since the time of George, economic language has become less and less adequate, mostly by becoming imprecise.

But, if you wish to use this mess, while casting doubt on a genuine attempt to provide our study with a scientifically based language, you may. You won't be wrong -- words used for defined concepts are never wrong. However, they may be suitable, or unsuitable, more or less useful, an aid to communication -- or not.

Other than myself in a paper to the AAAS, I know of only one outright attack on present terminology. This is by E. C. Harwood who founded the American Institute for Economic Research. In his book "Useful Economics" he analyzed 12 economics texts published between 1946 and 1954. They were used in a bunch of our most prestigious universities as well as University College, London.

He found half the books made no attempt to use the term "Wealth" precisely as the label for anything.

Of those who do use the term Wealth, three of them include land in the definition. In the case of "Labor", Harwood tells us in 3 of the 12 books "the authors use neither "labor" nor any other name to specify the human effort applied in processing things . . . "

In another five books, although the word "labor" is casually used, there is no attempt to use the word precisely for anything. That leaves the four who attempt precision. Their attempts arouse laughter. The landholder who makes the land available for use is a laborer -- but apparently not an entrepreneur -- who becomes a special kind of labor.

Look at this delightful statement the students at Dartmouth College had to read. They are told solemnly that the agents of production are three. Well, that looks promising. The three are (hold your breath) "Labor", "Capital including Land" and "Entrepreneurship".

Finally, the text used at the University of Virginia and Penn State asks delightfully (after describing Capital as "tangible, material goods" produced by labor from natural resources) "Is land Capital?" Well, they say, it depends on the usage of the label.

I won't continue with these conventional terms. I'll just let them lie there for your scientific use and pleasure.

Just one question -- can you see there could be a problem using "conventional terms" that may be deficient? Did you check and applaud or criticize their meanings?

Cliff Cobb wrote:

The Georgist notion that science consists of defining concepts, reasoning deductively, and providing a plausible explanation is a fundamental reason why this movement has had difficulty being taken seriously.

Response from Harry Pollard:

Apparently, you believe in a science that doesn't know what it is talking about -- because it hasn't made clear to itself what it means, whether they be words or mathematical symbols.

I think you have only used a mathematical symbol once in our discussion, for which I thank you.

Unfortunately, you used it badly. You said as you described my contentions:

Rent >> land value. ("When Rent is measured, invariably it is land value - a figure above and often way, way, above Rent.")

I didn't understand what you meant by "Rent >> land value" and said so.

You gave me a little homily on mathematical symbols. You said:

The symbol > means greater than. The symbol >> means much greater than.

Of course, I was wondering why, even though most of the paragraph pointed out that land value was usually "way, way, above Rent" you used mathematical symbols to say that "Rent is much greater than land value. Well, maybe you misused the math symbol -- something that happens to all of us. But please don't display an arrogance when referring to people on this list.

If you want empiricism, check into what the Pennsylvanians are doing -- particularly Dan, who is the ultimate empiricist as he deals with Pittsburgh councilmen.

Cliff Cobb wrote:

George himself did not seem to understand that science is about testable propositions, not plausible ones.

Response from Harry Pollard:

Plausible in its meaning of 'praiseworthy' -- or 'not to be trusted'? As our high school students enter the Cycles we spend a page or two on testable hypotheses and the prevalence of the untestable kinds. Maybe we are not Georgists in LA!

Cliff Cobb wrote:

In his discussion of the nature of science, he allows no consideration for methodologies for developing evidence. He seems to have believed that one could discover regular patterns (what he calls the "laws of nature") simply by direct observation and contemplation. It is a tribute to his genius that he derived so many propositions that have turned out to be true.

Response from Harry Pollard:

I would, at least, suggest the possibility that his true propositions might just owe their success to his methods.

Perhaps the major difference between science and philosophy is that science requires observation to substantiate its propositions.

Now, you agree with that (at least the bit about science) for in your next paragraph you make a point about statistics, which numbers depend on observation. You apparently believe that statistics are the truth. But with all due respect, the depend on observation.

I forget the English Lord who pointed out that all the figures counted depend at the bottom on the watchman on his beat. (If anyone can tell me, I would be obliged.)

Those magnificent measurements you love so much have gone through so many levels on their way to the top, picking up errors - and later, subjective corrections - I cannot see why you would prefer them to observations at the first level with little error and easy testing.