A system to categorize various design options for brains according to their ability to react to a given situation. Proposed by philosopher Daniel C. Dennett in 1975, and very similar to an earlier idea by Konrad Lorenz.

The first floor of the Tower is made up of Darwinian creatures, named for Charles Darwin's theory of natural selection. These creatures have various genotypes, {W, X, Y, Z}, that code for phenotypes {W, X, Y, Z}. The phenotypes are variations on one biological model When confronted with situation Q in its environment, each phenotype reacts differently. Phenotypes W, X, and Z tend to react to situation Q in a way that will keep the creature with the corresponding genotype from being able to reproduce, by killing the creature or otherwise ending its "family tree" as a result of situation Q. Only phenotype Y tends to allow survival/reproduction, which means that genotype Y will soon outnumber W, X, and Z in the environment. Darwinian creatures are "hard-wired" and can only react in the narrow ways their phenotype will allow. Individual creatures are unable to choose the options available, and are thus largely insignificant to evolution on this floor.

The next floor is made up of Skinnerian creatures, named for B.F. Skinner and his school of psychology known as behaviorism. These creatures have a phenotype that can offer options W, X, Y, and Z when confronted with situation Q. They are also equipped with an ability to recognize positive and negative input from their environment, but otherwise "blindly" try random options when confronted with situation Q. Options W, X, and Z tend to cause negative input from the environment, while option Y tends to cause positive input. Eventually, the individual creature will tend to use option Y when confronted with situation Q. B.F. Skinner's behaviorism is the theory that all animals, from sea slugs to people, are these types of creatures. Dennett states, "Skinnerian conditioning is a fine capacity to have, so long as you are not killed by one of your early errors". That is, individual creatures can eventually become more fit for their environments, as long as they don't choose option X first, which leads to immediate death or failure to reproduce.

The third floor contains creatures with a system of preselection, a mental ability to weed out the "truly stupid" options before attempting them in the real world - that is, ruling out options X and Z altogether when first encountering situation Q, because the creature's mind can foresee bad consequences without prior conditioning for that exact situation. These creatures are called Popperian, for Sir Karl Popper, who once said that this design element "allows our hypotheses to die in our stead". This will make the creature more likely to choose option Y immediately, when first facing situation Q, and thus survive and reproduce. Fit Popperian creatures can be differentiated from fit Skinnerian creatures: the former are smart enough to make better-than-chance first moves, while the latter are only lucky enough to make the right first moves.

Popperian creatures, in Dennett's view, make up most of the animals used in behavioral studies: they have an inner environment, however primitive, that allows them to preselect options. Dennett notes that the sea slug has replaced the pigeon in Skinnerian studies, as pigeons (Skinner's favorites) and other creatures that are not simple invertebrates can be shown to exhibit some preselection. Even the octopus has been shown to be an astoundingly smart Popperian creature by its ability for preselection. But is there a separation of degree between humans and other Popperian creatures? After all, humans do their share of mindless preselection. But Popperian creatures have limits that go back to their biological origins: that is, if Darwinian natural selection has made options Y and Z "unthinkable" to the creatures because of their genetic nature, then Popperian preselection cannot consider those options. It is on this level that psycholinguist Noam Chomsky places human thought, hypothesizing that there are certain things that humans cannot know, in the same way that fish cannot understand what we call "fishing", much less democracy. Chomsky has theorized that perhaps finding the answer to the problem of free will is such an option, ruled out by our biology and prior selection.

Dennett disagrees. He theorizes yet another level of creatures in the Tower, the Gregorian, named for the British psychologist Richard Gregory, who is known for his theory of Potential Intelligence and Kinetic Intelligence. Gregory theorizes on the role of information in making "smart moves" the first time. For instance, he examines a pair of scissors - a well-designed artifact - and notes that this is not just a result of intelligence, but something that endows its user with Potential Intelligence. That is, a person with a pair of scissors is more likely to finish the task "divide this piece of paper in two" swiftly and safely than a person without scissors. Tool use has always been considered a mark of high relative intelligence. Chimpanzees use twigs to take termites from termite mounds, thus opening up a food source and allowing greater chance for survival as a species. But even more notable is that there are groups of chimpanzees who cannot "fish" for termites when they are present as a food source - they never "learned". This allows the classification of two chimpanzee "cultures". Twig-use doesn't only require intelligence, but confers it as well, as comparing two individual chimpanzees from different "cultures" will show us. Thus, chimpanzees are a Gregorian creature: they can absorb outer-environment features into their own inner environments.

Humans are also Gregorian creatures, albeit ones with a larger environment to work with because of our greater capacity for creating and using tools. Perhaps the most important of our tools are words and concepts: "teamwork" and "inclined plane" are among the tools that we can apply to various situations, as well as higher concepts like "president", "love", "node" and "ska-core", and specific instances like "Everything2", "Chinese New Year", and "Dennett". Darwin (1871, p. 57) posited this trait, that of language, as the prerequisite for "long trains of thought" and thus the human ability to alter its environment in a way umatched by other species. These theories have been supported in recent years by Julian Jaynes (1976) and Howard Margolis (1987), who suggest that self-exhortations and reminders available to us as language-users are what allow us to, say, build dams better than beavers even though we are not born with a complex biological instinct to do so like beavers are.

Dennett posits the final level of the tower as that at which humans have arrived: that of science. This is a system that we have created using Gregorian methods of foresight and earlier methods of trial-and-error to embody a better system of generate-and-test that is less arbitrary than all the rest. Situations can be fed into this singular system and accurate results can be gathered without having to actually encounter the situation firsthand - what is known as a hypothesis. Science is also not just a matter of making mistakes, but of making mistakes in public - that is, the human capacity for language allows for a single individual to build upon the results of earlier scientific tests by proposing new experiments for new hypotheses. Thus, there is something of a new genetics in the science of ideas, where old ideas that do not work, like the Ptolemaic theories, do not need to be theorized by each individual and subsequently demolished in order to be replaced by more relevant ideas as the prevailing understanding of a given situation (in this case, the solar system). This makes naked animal Popperian brains no match for our heavily-equipped scientific society, and reverses the burden of proof on Chomsky and other theorists who posit "cognitive closure" (the aforementioned theory that human brains are simply closed to consideration of certain options, even when they are "obvious"). Science provides for a new way of solving problems that can encompass things outside of the Gregorian mien we are born into.

(An interesting project would be a writeup of how this theory applies to Everything2 and its noders.)


Chomsky, Noam. 1972. Language and Mind. Enlarged ed. New York: Harcourt Brace Jovanovich.

Dennett, Daniel C. 1975. "Why the Law of Effect Will Not Go Away." Journal of the Theory of Social Behaviour, vol. 5, pp. 179-87. Reprinted in Dennett 1978.

Dennett, Daniel C. 1995. Darwin's Dangerous Idea. Touchstone ed. New York: Simon & Schuster.

The term is capitalized by Dennett thusly: "Tower of Generate-and-Test". The gods that be have decided that it is best noded as above.

Log in or register to write something here or to contact authors.