SETI: A Critical History: 4. The Rehabilitation of 'Mind'
Chapter 4 of the thesis
How the Search for Extraterrestrial Intelligence Became Disconnected from New Ideas About Extraterrestrials
Mark A. Sheridan
We were not out to reform behaviorism, but to
– J. Bruner
In this chapter we briefly explore a major development in the field of psychology that had important consequences for the ETI discourse and thus, ultimately, the SETI project. SETI's architects came of age intellectually at a time when behaviorist psychologists considered the concept of "mind" to be embarrassingly unscientific. As such, the SETI scientists probably spent little – if any – time thinking about the possible "nature" of ETI as they designed their project. Instead, they made the default assumption of the traditional ETI discourse: that an ETI had a humanoid cognitive structure, consciousness, and intelligence.
Ironically, at about the same time SETI got underway experimental psychologists began to embrace assumptions and methods that brought about major changes in their field. The vocabulary of cognitive science, as the emerging field came to be known, enabled scientists to discuss human cognition, consciousness, and intelligence with a new legitimacy. This, in turn, profoundly changed the ETI discourse. For the first time the discourse became self-consciously aware of its traditional assumption of humanoid ETIs. Moreover, contributors to the discourse used the new tools of cognitive science to imagine non-humanoid ETIs. The ETI discourse expanded dramatically as the new question about the possible nature of ETI joined the traditional question of whether ETIs exist. As the discourse expanded, SETI became grew increasingly disconnected from it. For the first time critics noted SETI's traditional assumption of humanoid ETIs and to imagine the difficulties SETI might have understanding, or even detecting, messages from the new, non-humanoid ETIs.
At the beginning of the 20th century many psychologists felt their field did not enjoy the same institutional status given to "hard" sciences. They believed this to be the result of psychology's methodology – it relied heavily on the subjective technique of introspection – and the fact that psychological theories included non-empirical constructs such as mind and consciousness. By 1913 John B. Watson thought he and his colleagues had restructured the field in a way that addressed these issues. In his essay "Psychology as the Behaviorist Views It" Watson proclaimed the field's new mission to be the prediction and control of behavior.82 Behaviorism studied stimuli and the responses elicited by those stimuli; behavioral psychologists ignored what went on inside the mind. Behaviorists now claimed their empirical foundation to be as "hard" as that of any science. According to Thomas Leahy, a historian of psychology, "by 1930, behaviorism was well established as the dominant viewpoint in experimental psychology."83
During the 1930s, when the oldest of SETI's architects commenced their undergraduate studies, B. F. Skinner's "radical behaviorism" began to gain favor among psychologists. It was radical because it not only chose to ignore the role an individual played in intermediating stimuli and responses, but went so far as to say that an individual had no role: one's environment determined how individuals behave.84
SETI's architects and the behaviorists were kindred spirits, in that both sought to establish a field of inquiry on rigorously scientific grounds and thereby gain more respectability among scientists. Ironically, by the time the SETI project crystallized psychology had already entered yet another period of dramatic transformation. Over a twenty-year period just prior to Drake's first search, a number of brilliant insights each exerted an enormous influence in its field. Together, these insights enabled psychologists to discuss concepts like cognition, consciousness, and intelligence with a previously unattainable degree of scientific rigor. The concept of mind was rehabilitated from the dustbin to which the behaviorists had proudly dispatched it. The dominant paradigm in experimental psychology shifted from behaviorism to what we today call cognitive science, in which behaviorism plays a relatively minor role. An early cognitive scientist, Jerome Bruner, recalled "what I and my friends thought the revolution was about back there in the late 1950s.... We were not out to 'reform' behaviorism, but to replace it."85
This is not the place to attempt even a brief history of these important changes. But I do want to identify some of the milestones in this transformation because the insights that underwrote this change also made possible the new thinking about the nature of ETI that transformed the ETI discourse. Once scientists identified a vocabulary that could describe a being's nature in ways the scientific community considered legitimate, they used it to articulate ideas about how ETI's nature might differ from that of humans.
The timing of these transformational events is particularly relevant. The conceptual ground was shifting under the SETI pioneers' feet even as they designed their project. The Soviets enjoyed the benefit of just a few additional years before they commenced CETI, and the extra time afforded them better access to the new tools of cognitive science.86
One obvious starting point of this transformation was the claim by Alan Turing in 1936 that a machine that could perform logical calculations, a property previously attributed exclusively to the human mind.87 Then, in a 1943 article, the neurophysiologist Warren McCulloch and the logician Walter Pitts formalized the analogy between a computer and the human mind. They "showed that the operations of a nerve cell and its connections with other nerve cells (a so-called neural network) could be modeled in terms of logic. Nerves could be thought of as logical statements, and the all-or-none property of nerves firing (or not firing) could be compared to the operation of the propositional calculus, where a statement is either true or false."88 John von Neumann's description around the same time of stored computer programs provided the last important idea needed to complete the metaphorical link between the human mind and a computer.89
Also in 1943, Norbert Weiner and two colleagues published the first paper on cybernetics. The core of cybernetics was a very specific kind of information called feedback. A simple example is a thermostat. Information about the actual temperature is compared, in a feedback loop, to the desired temperature, and some kind of purposive behavior results. If, say, a room is colder than the desired temperature, that information causes the heater to turn on; when the room reaches the desired temperature, information about that state causes the heater to turn off. In a sense, the heater "learns" when to turn itself on and off. The feedback loops of cybernetics that produced purposive behavior represented another way to explain the working of the human mind in terms of a concrete machine process rather than the abstract introspective and philosophical notions of mind that the behaviorists wanted to discard. Cyberneticists saw machines and humans – and many other entities, like economies, societies, and ecosystems – as self-regulated systems.90
Once computers, the computer metaphor of mind, and self-regulating systems that could learn appeared, the idea of artificial intelligence soon followed. In 1950 Turing published a paper, "Computing Machinery and Intelligence," that is widely credited with defining the field of artificial intelligence by specifying what behaviors a machine would have to exhibit in order to be considered intelligent. Cognitive scientists turned this insight around, using Turing's paper to support their use of the computer as a metaphor for the brain.91
Claude Shannon's formulation of "information theory" was arguably the most critical step in the transformation. Like many of the pioneers of cognitive science Shannon formulated his ideas while working on military projects during World War II. It is hard to overestimate the impact Shannon had; the Digital Age would not have happened until someone had Shannon's insight that information could be reduced to binary digits (bits).92 The significance of information theory for cognitive science lies in the fact that information can be treated like a physical quantity – it can be measured and modeled – without having to also explain its semantic content. Information can be experimentally studied in a way that concepts like consciousness could not.
In 1956, the year of the Dartmouth conference on artificial intelligence, George Miller published an early paper on the phenomenon now referred to as chunking.93 The AI pioneer Marvin Minsky once calculated that a being looking at a scene containing fifty objects would be able to assemble them into a quadrillion different sets.94 Because such a vast amount of information from both its external and internal environments bombards any sentient being, it evolves filters that attend selectively to it. Beings also evolve processes that impose an order onto whatever information does make it through these filters. This order is a kind of fiction that the being constructs because doing so confers evolutionary advantage. We normally think of these narratives constructed by a being about its environment and its self as external and internal realities. The significance, insofar as SETI is concerned, lies in the fact that different beings chunk the available information differently, because they have had different evolutionary histories. In other words, different beings construct realities differently.
The year before Cocconi and Morrison published their paper, the AI pioneers Allen Newell and Herbert Simon published a paper arguing that a psychological theory, in an obvious departure from behaviorism, "should explain ... what processes are used and what mechanisms perform these processes." Also, in their theory "an explanation of an observed behavior of the organism is provided by a program of primitive information process that generates this behavior." Mind was once again an object of legitimate study, and the computer's ability to process information served as the preferred model for it.95
Finally, in 1959, the year of Cocconi's and Morrison's paper, Noam Chomsky set the field of linguistics on a new path. In a review of Skinner's book Verbal Behavior Chomsky exposed a number of weaknesses in the behaviorist model in what the historian Leahy called "perhaps the most influential paper in the history of psychology since Watson's manifesto for behaviorism."96 Drawing on theories he developed, Chomsky demonstrated behaviorism's inability to explain the human use of language exclusively in terms of stimuli and responses. Instead, Chomsky showed the necessity of internal processes – mind – in the form of a finite number of rules than could generate an infinite number of intelligible sentences. Chomsky abandoned linguistics' traditional focus on meaning and substituted "information" and "syntax." In Chomsky's linguistics the human mind performed information-processing and symbol-manipulating functions analogous to those performed by a computer. Language, thought by many to be the vehicle in which the mind's rationality is embodied and made manifest, was thus the product of physical components performing describable processes, just like a computer.97
Later Chomsky took his theory an additional step farther away from the behaviorists' position that language was simply one more learned behavior, and an additional step toward the rehabilitation of mind. In Language and Mind he argued that language was a biological system. Evolution endowed us with a universal grammar that allowed us to communicate with any healthy member of the species. An analysis of language behavior could not yield an understanding of language, any more than watching a bird fly could produce an understanding of lift and the other principles of aerodynamics that explain flight. "The major contribution of the study of language will lie in the understanding it can provide as to the character of mental processes and the structures they form and manipulate." Linguists, too, had joined the ranks of cognitive scientists who studied mind from the vantage point of their field.98
The new field of cognitive science focused attention on ways that humans constructed reality using a cognitive structure, consciousness, and intelligence – a nature – produced by evolution. In the remainder of Part II we examine how the ETI discourse quickly picked up these insights. The possibility that other intelligent beings may have evolved natures that differed significantly from ours suggested to critics of SETI that ETIs might construct messages that SETI would find difficult or impossible to understand or even detect.
B. Watson, "Psychology as the Behaviorist Views It," Psychological
Review 20: 158–77.
83. Thomas H. Leahy, A History of Psychology – Main Currents in Psychological Thought, 6th ed. (Upper Saddle River, NJ: Pearson/Prentice Hall, 2004), 377. See Chapters 11–12 for a robust history of behaviorism.
84. Morrison was born in 1915 and was working on the Manhattan Project as a freshly-minted Ph.D. in 1945. Oliver was born in 1916, received his Ph.D. in 1940, and then went to work for Bell Labs during the war; he joined Hewlett Packard in 1952 as the Director of Research. Drake and Billingham were both born in 1930. Drake entered Cornell in 1949, interrupted his studies to join the military, and got his Ph.D. from Harvard in 1958. Billingham, an Englishman, was a physician in the Royal Air Force from 1956 to 1963, when he joined NASA. Sagan was born in 1934 and entered the University of Chicago in 1951.
85. Jerome Bruner, Acts of Meaning (Cambridge: Harvard University Press, 1990), 3.
86. There are many good histories and summaries of the advances made in cognitive science. For a concise introduction see John Tooby and Leda Cosmides, "Forward," Simon Baron-Cohen, Mindblindness – An Essay on Autism and Theory of Mind (Cambridge, MA: The MIT Press, 1995). A standard textbook is Michael S. Gazzaniga, Richard B. Ivry, George R. Mangun, Cognitive Neuroscience – The Biology of the Mind, 2nd ed. (New York: W.W. Norton, 2002). Two historical treatments are Howard Gardner, The Mind's New Science – A History of the Cognitive Revolution (New York: Basic Books, 1985); and Brigitte Chamak, "The Emergence of Cognitive Science in France: A Comparison with the USA," Social Studies of Science 29 (October 1999): 643–84. The epistemological issues of most relevance to the ETI discourse are discussed in Patricia Smith Churchland, Brain-Wise – Studies in Neurophilosophy (Cambridge: The MIT Press, 2002). A popular text is Steven Pinker, How the Mind Works (New York: W.W. Norton, 1997).
87. Alan Turing, "On Computable Numbers, with an Application to the Entscheidungsproblem," Proceedings of the London Mathematical Society, Series 2, 42 (1936): 230–65.
88. Warren McCulloch and Walter Pitts, "A Logical Calculus of Ideas Immanent in Nervous Activity," Bulletin of Mathematical Biophysics 5: 115-33. For a discussion of their insight see Gardner, 18.
89. John von Neumann, "First Draft of a Report on EDVAC," at www.virtualtravelog.net/entries/2003-08-TheFirstDraft.pdf, accessed 8 August 07.
o 90. Arturo Rosenblueth, Norbert Weiner, and Julian Bigelow, "Behavior, Purpose, and Teleology," Philosophy of Science 10 (1943): 18–24. See Also Norbert Weiner, Cybernetics: Or the Control and Communication in the Animal and the Machine (Cambridge: MIT Press, 1948).
91. A. M. Turing, "Computing Machinery and Intelligence," Mind 49: 433–60. For a discussion of this insight see Leahy, 420.
92. Claude E. Shannon and Warren Weaver, The Mathematical Theory of Communication (Urbana, IL: University of Illinois Press, 1949. For a less technical introduction to the topic see John R. Pierce, An Introduction to Information Theory – Symbols, Signals & Noise, 2nd ed. (New York: Dover, 1980).
93. George Miller, "The Magical Number Seven, Plus or Minus Two: Some Limits on our Capacity for Processing Information," Psychological Review 63 (1956): 81–97.
94. Marvin Minsky, "Communication with Alien Intelligence – It May Not Be as Difficult as You Would Think," BYTE (April 1985): 134.
95. Allen Newell, Herbert Simon, and J. C. Shaw, "Elements of a theory of problem solving," Psychological Review 65 (1958): 151.
96. Leahy, 407.
97. Noam Chomsky, Syntactic Structures (The Hague: Mouton, 1957).
98. Noam Chomsky, Language and Mind (NY: Harcourt Brace Jovanovich, 1968), 58.