I collect star atlases and catalogs. I don’t specialize in antiques or high-end relics of unusual artistic or historical value, I prefer out-of-print items that are still useful to amateur observers because they present convenient, one-stop accumulations of real data no longer available without extensive and tedious research of on-line databases.

All astronomical references have errors or omissions such as typos, bad observations, or misidentified objects. These flaws are often copied faithfully onto new editions and compilations (no editor wants to inadvertently discard questionable entries that might be valid and thereby be clues to genuine phenomena. Old references, even if they are in error, are useful in helping evaluate the historical evolution of a fluid body of knowlege and tracking down when and where errors entered the canon so they can be positively identified. This sort of cross-checking is vital because even revised or corrected references often leave out legitimate entries if an editor mistakenly felt they were incorrect. Incidentally, our on-line reference databases are also plagued with these problems, so you still need the old books to help sort out the mess.

As a result, there is a brisk trade in these items, mostly by amateurs, since the pros don’t have time or resources for this meticulous, thankless, (and frankly, boring!) work. Lately, I and my fellow Bartlebys in this business have noted a strange new development. The new, on-line book vendors like Amazon have been charging outrageous prices for some out-of-print, but still readily available volumes. And the prices rise quickly, often on a daily basis. The demand just isn’t there to justify these prices, and those of us in the business suspect it is an artifact of the marketing process, not a true supply and demand issue.

In the past, these old books could be picked up for a song at estate sales, library clearances, or publisher house-cleanings. Some of these books are not even particularly old, to quote just one example, the Millennium Star Atlas was overpriced at several hundred dollars, so publication was discontinued because it just wasn’t selling well. It was a wonderful reference and well reviewed, but more reasonably priced alternatives with comparable features were readily available to observers. It is, after all, a free market, and that’s how free markets work. Today it is being sold at some on-line outlets for close to a thousand dollars. A paperback edition of the same book was soon printed, but even at a substantial discount, the market was limited and it went out of print as well. It is still possible to pick up this jewel for close to its original bookstand price, but you have to be patient, know the market, and keep up with the other amateur uranographer’s websites.

Similar phenomena have been noted with other popular works, like the Becvar atlases, The Harald-Bobroff Astroatlas and even the new Great Atlas. The latter is a massive, monumental work assembled from the best satellite data and catalogues; four million stars and 70,000 non-stellar objects, and the publishers are discounting it down to less than half its modest original price. They can’t give it away. These are all scientifically worthwhile documents, and they each possess useful features the others lack. They complement each other nicely, and amateurs would love to buy all of them at their original prices, or near them! I did.

This price spike in on-line outlets could be explained as artificial demand generated by speculators, the sort of thing we’ve seen in the 17th century Dutch tulip market or the 21st century US real estate boom, but many of my fellow uranographers disagree. They suspect it is an artifact of automated programs, bots, which monitor other online retailers and force prices higher than existing quotes in the hope someone unfamiliar with the market will take the hook. Real markets shouldn’t work this way, prices should follow demand, not lead it. Now that much equity trading is done by software, not by people, we’ve seen the phenomenon in stocks. Its counter-intuitive. Adam Smith argued persuasively that a large human market informed by instant communications should drive prices down, not up.and they should quickly stabilize at a level both buyer and seller could live with. After all, no one actually NEEDS these books to do research. They are all available at university libraries, and anyone knowlegeable enough to want them is perfectly aware of that. This isn’t the art market, these prices are not driven by speculators or unscrupulous financiers gaming the system.

Or, as some of my correspondents suggest, what we’re seeng is flaws in classical economic theory manifesting themselves under the extreme new conditions brought on by technology. The bots want to sell at the highest possible price, and unlike human sellers, are willing to bet demand will go up to meet the price, instead of the other way round. And as software takes up more and more of our economic decision making, this behavior could spread to other, more vital markets. This could be the crucial experiment that demonstrates classical capitalistic theory was wrong, just as minute, but measurable and undeniable anomalies in Mercury’s orbit demonstrated that Newtonian Gravitation needed to be replaced by General Relativity. A breakdown, even in extreme cases, of the supply-and-demand paradigm would be earth shaking. This is, after all, the foundation of all our economic thought. Even Karl Marx trusted heavily in, and relied completely on, Smith’s Wealth of Nations to formulate his theories. He argued capitalism would collapse because it inevitably would try to ignore market forces and try to avoid the consequences of doing so . If we really don’t understand how markets work, then not only is capitalism in trouble, but Socialism is too.