There's a fascinating article in this week's New Yorker by Malcom Gladwell, called "In The Air," that makes a compelling case that scientific discoveries very frequently occur in multiples, i.e. instances where several scientists independently come to the same discovery. Examples include Newton's and Leibniz's independent discover of calculus, and the independent invention of the telephone by Alexander Graham Bell and Elisha Gray. An interesting corollary of this observation is that eponymous discoveries are also very frequently named for the wrong person. From Gladwell's article:

The statistician Stephen Stigler once wrote an elegant essay about the futility of the practice of eponymy in science that is, the practice of naming a scientific discovery after its inventor. That's another idea inappropriately borrowed from the cultural realm. As Stigler pointed out, "It can be found that Laplace employed Fourier Transforms in print before Fourier published on the topic, that Lagrange presented Laplace Transforms before Laplace began his scientific career, that Poisson published the Cauchy distribution in 1824, twenty-nine years before Cauchy touched on it in an incidental manner, and that Bienaymé stated and proved the Chebychev Inequality a decade before and in greater generality than Chebychev's first work on the topic." For that matter, the Pythagorean theorem was known before Pythagoras; Gaussian distributions were not discovered by Gauss. The examples were so legion that Stigler declared the existence of Stigler's Law: "No scientific discovery is named after its original discoverer." There are just too many people with an equal shot at those ideas floating out there in the ether. We think we're pinning medals on heroes. In fact, we're pinning tails on donkeys.

Stigler's Law was true, Stigler gleefully pointed out, even of Stigler's Law itself. The idea that credit does not align with discovery, he reveals at the very end of his essay, was in fact first put forth by Merton. "We may expect," Stigler concluded, "that in years to come, Robert K. Merton, and his colleagues and students, will provide us with answers to these and other questions regarding eponymy, completing what, but for the Law, would be called the Merton Theory of the reward system of science."

I certainly came away from the article believing even more strongly in the Boldrin-Levine contention that intellectual property rights just aren't necessary when you have the shoulders of giants to stand on.

Last week's In our Time Radio 4 BBC weekly program on the Translation Movement (http://www.bbc.co.uk/radio4/history/inourtime/inourtime.shtml) should be of interest to every economic historian, and especially those interested in the role of intellectual monopoly (property) in the growth of knowledge. That is, there was none - ie no intellectual property - and two breathtaking renaissances with out it, one Arabic, the other European. According to Melvyn Bragg and his 3 commentators the 9th century translation of Greek texts on health, mathematics, philosophy, religion, and engineering into Arabic was THE critical factor in the flowering of science and culture in the Islam world for the next several centuries , and indirectly (through the much later translation of Arabic texts into Latin) was hugely influential in the growth of the European Renaissance. Of course we're also talking "derivative works" here, since "translation" often involved detailed commentary, elaboration and interpretation, and what the author regarded as correction, as well as modification. Do we have a "natural experiment" here? Two great eras of flourishing knowledge in science, philosophy, arts, trade commerce ... ie culture ...built on, amongst other things, absence of tightly enforced national or international exclusive rights in knowledge products ? Over at my blog on strategicecon.com (http://strategicecon.com/post/1/172) I explore this idea a little further ....and conclude: "I guess it is a sign of our times that we regard translators who don't pay as "pirates" and prosecute them as criminals , whereas during these Arabic and European "Renaissance" era transaltors who didn't ask permission or pay fees for their oriignals were regarded as cultural heroes (and well paid for it) ." PS I couldn't get your link tool working in the editor in Firefox...)

Amazing the in-depth scientific bases used in these posts. By the comments above, we should not be developing medicines because people have gotten well without them. Indeed, why do we have cars or planes? People were able to get from one place to another without them for millenia. The list goes on and on. We did we bother to develop cell phones? People were quite happy to communicate through other means prior to the invention of cell phones.

However, there is one theme behind all these things; in every instance, things happened faster, or more efficiently.

So, what is the lesson for intellectual property? Speed and efficiency are again advantages. If ten individual inventors invent ten things, and only one is commercialized and the other nine inventors decide they are going to do nothing with their invention for whatever reason, without a system of patents those nine inventions are lost to the world until someone else does the same work all over again (loss of speed and inefficiency have both occurred). Further, the tenth inventor's invention may well be something he uses in Butte, Montana, but people needing that invention in Florida, Germany and Brazil are out of luck because no one sees the invention until seven years after it was invented.

Another point: What if there had been a formal intellectual property system during the two eras of scientific discovery, advancement, etc., noted by John Fountain? I wonder whether the eras, which lasted for centuries, might have been compressed because knowledge would have been located in a single location where everyone could have accessed it instead of being in esoteric locations with access limited to a few.

Essentially, what meaning does intellectual property have when the time frames begin with decades and extend into centuries? Little or none. Further, as Stephen Spear noted in his post, all these lovely discoveries by Pythagoras, Chebychev and others were mathematical principles that had no physical embodiment at that time; no intellectual property (in terms of patents) here. Further, copyright does not apply because mathematical equations are what they are and in and of themselves are not copyrightable. Even today, academics share knowledge back and forth freely, but the vast majority of that knowledge in math and physics does not meet the requirements of intellectual property.

John Fountain's post is hilarious. On the one hand he notes that the translations from Greek formed the basis of progress for centuries, and then he turns around and says that it is humorous that there were no intellectual property laws. Okay, what if there were intellectual property laws? The translated texts were almost all centuries old, well past any intellectual property protections, even those in modern times and even the ridiculous lengths for copyright set by the United States.

While interesting, the "natural experiment" is merely an observation of what happened over a period of centuries, and is neither more or less relevant than what has happened in any other set of centuries up to the present day. There may be lessons to be learned, but the technology and situations are different, and our response will surely be different as well.

Read the MPEP (the guiding document of the USPTO). It says in so many words that the vast majority, and perhaps all, inventions rely on the works of earlier inventors, as they almost assuredly must. Does that make the invention of the catalytic converter or FM radio any less startling, or easy? What about the substantial work that went into making the first microwave oven? Rather than implying that inventorship is easy because there is so much history to build upon, I recommend reading about the years and even decades of work some inventions took before belittling the effort it takes to invent.

Another point:

I believe each of the items mentioned in Stephen Spear's original post were "discoveries," rather than "inventions." The patent act specifically precludes laws of nature, physical phenomena and abstract ideas from being patentable matter. Thus, Gladwell's characterization that discoveries of mathematical relationships, which in general are not patentable for a variety of reasons, are "inventions," is incorrect. In the words of the USPTO, these equations are not statutory patentable matter.

I also did a bit of casting about with respect to names. The Pythagorean theorem was named centuries after the Pythagorean school existed, hardly the fault of Pythagoras. However, Pythagoras (or his school) was the first to "popularize" the theorem and the first to express it mathematically. On the other hand, Pythagoras believed in trade secrets and his society held some things secret for 200 years - even killing a student who dared to utter one of their secrets in public. Ah, what wild and woolly days before intellectual property!

Gauss's name was likely used for the Gaussian distribution because of his defense of the distribution in 1809. I am unsure of whether Gauss claimed to be the discoverer of the distribution.

With respect to Chebyshev, many people used the name Bienayme-Chebyshev Inequality because Bienayme proved the inequality substantially earlier than Chebyshev (who was good friends with Bienayme) did. However, according to Markov, a student of Chebyshev, Bienayme proved the inequality because of an argument he had with Cauchy. Chebyshev, again according to Markov, understood the purpose and value of the inequality, and deserved to have his name applied to the theorem (See the trend here? Frequently discoverers have nothing to do with naming, which is how we end up with oxymorons).