Inventing Invention: A Case Study of Legal Innovation

Intellectual Property and Patents

Article Snapshot


John Duffy


Texas Law Review, Vol. 86, pp. 1-72, 2007


This paper relates how ideas about what should be patentable have changed over time.

Policy Relevance

Innovation in law happens slowly and requires courts that are willing to disagree with other courts.

Main Points

  • Today, an idea must be new, useful and not obvious in order to be patentable in the United States. Early patent systems did not ask whether an invention was obvious, as technology development was spare and slow, so no one worried there could be too many patents.

  • The history of the U.S. patent system offers examples of legal innovation. Legal innovation is very slow and many new ideas fail.

  • Amazon’s patent on a one-click method of ordering items online might be too obvious to be patentable. Several other developers came up with similar ideas at the same time, a red flag for obviousness.

  • Obviousness ideas helps keep a patent narrow; so the patent protects just one kind of better engine, not all “faster engines.”

  • Centuries ago, early patents were developed in Venice and then in England, but the law moved very slowly and unevenly towards an objective test for whether an invention is worth patenting.

  • Early tests for whether an invention is worth patenting tend to be subjective, like the “flash of genius” test, and could be vague. Later tests, like the Federal Circuit’s teaching-suggestion-motivation test, were too rigid and were rejected by the Supreme Court.

Get The Article

Find the full article online

Search for Full Article