Artificial Intelligence and Natural Man
Boden (Margaret)
This Page provides (where held) the Abstract of the above Book and those of all the Papers contained in it.
Colour-ConventionsDisclaimerBooks / Papers Citing this BookNotes Citing this Book


Back Cover Blurb

  1. Artificial Intelligence and Natural Man is the best introduction to artificial intelligence available for general as well as professional readers.
  2. This revised edition adds a chapter and significantly extends the bibliography. Among the new topics addressed are
    → The impact of Japan’s Fifth Generation project on artificial intelligence research worldwide,
    → Advances in low-level vision,
    → The new connectionists,
    → Boltzmann machines,
    → Logic programming,
    → Robotics,
    → Expert systems and learning programs, and
    → The role of Al in military technology and the Strategic Defense Initiative.
  3. “The best book I have ever seen on nearly all aspects of artificial intelligence, including technical questions, philosophical questions, and the like. It is a rich book, and in my opinion, a classic.”
    Douglas Hofstadter, author of "Hofstadter (Douglas) - Godel, Escher, Bach: an Eternal Golden Braid - A Metaphorical Fugue on Minds and Machines in the Spirit of Lewis Carroll"
  4. ‘This is the human being’s guide to machine intelligence. . . . [Boden’s] grasp of the whole area is formidable, and her background in psychology and philosophy guarantees that the study is deep as well as broad.”
    → Audrey Laski, Times Educational Supplement
  5. "This was, and still is, a remarkable book: nothing else then, or since, both surveys a wide area of work in artificial intelligence and performs a philosophical and analytical synthesis of the field.”
    Yorick Wilks, Director of the Computer Research laboratory, New Mexico State University

Preface To Second Edition
  1. Artificial Intelligence has grown dramatically since this book first appeared: there are new researchers — and research groups — throughout the industrialized world. Its visibility has increased also, for the media have now discovered it.
  2. But growth, whether visible or invisible, can take place without development. The recent explosion of funding and publicity is due to commercial and political factors, not to intellectual advances within the field. The central problems of artificial intelligence, and the theoretical basis of its achievements, have remained essentially the same. Naturally, there has been some advance, but most of it is in technological efficiency, not basic scientific understanding.
  3. Despite the passage of time, then, the main message of this book is still timely. The book is not a research text on the nitty-gritty of programming, nor a "list" of the most up-to-date programs around. Indeed, two opening chapters (Chapters 2 and 3) describe a program (a simulation of neurosis) that was already out-of-date when I started writing the first edition in the early 1970s. I chose to discuss it nevertheless — and would do so again today—because its simplicity, and its obvious "stupidity," help introduce two wide-ranging questions: What sort of thing is a computer program? and What difficulties face those who try to make programs as powerful as human thinking? These questions are what this book is about, and they are hardly nearer an answer now than they were a decade ago.
  4. To be sure, if I were to write this book "from scratch" now, it would not be exactly the same. Some recent programs would be included in addition to — occasionally, instead of — the ones originally chosen. And many projects cited in the first edition as exploratory research have been further developed, so could now be described in more detail.
  5. In particular, I would now devote an entire chapter to low-level vision, instead of discussing it within the chapter on psychology (Chapter 13). Low-level vision is the area in which there has been the most significant theoretical advance. I would include an account of the current exploratory research on parallel processing ("connectionist") systems. I would allow more space to "production rules," because of their use in commercially available expert systems like those mentioned in Chapter 15. And in view of the current interest in logic programming, I would add PROLOG to the programming languages discussed in Chapter 12. (I discuss the first three of these topics at length, though from the psychologist's — not the technologist's — point of view, in Boden [in preparation].)
  6. Comparatively minor adjustments would be made also. For example, my discussion of creativity (Chapter 11) would now include programs on heuristic exploration, which have been used to help plan experiments in genetic engineering and design three-dimensional "chips." Chapter 10 would refer to a recent theoretical analysis of learning programs that highlights their computational similarities and differences. Further work on text analysis and story understanding has been done by the workers mentioned in Chapters 7 and 11. And in Chapter 7, too, I would discuss attempts to program conversation, which previously I mentioned only in the notes. But none of this new material would invalidate the essential points made in the original version.
  7. For this second edition, then, I have not altered the main text1. Instead, I have added a Postscript chapter, and a separate Bibliography listing significant recent work. The Postscript explains the reasons for the recent upsurge of publicity, indicates the areas of theoretical novelty, identifies the main current research issues, and suggests what advances we can — or cannot — expect in the future.
    → M.A.B., Brighton, Sussex, April 1986

    Preface To Second Edition – xi
    Preface To First Edition – xii
    Acknowledgments – xiv
  1. Introduction
    1. What Is Artificial Intelligence? – 3
      → Programs Tell Computers What To Do
      → Languages The Computer Understands
      → Computers Manipulate Symbols
  2. The Personal Dimension
    1. Artificial Neurosis – 21
      → Outline Description Of A Neurotic Program
      → Defense Mechanisms That Distort Beliefs
      → Later Improvements
    2. Function And Failure In The Neurotic Program – 34
      → Data Used By The Neurotic Program
      → How The Neurotic Program Works
      → Failings Of The Neurotic Program
    3. Personal Politics And Ideology Machines – 64
      → Simulation Of Hot Cognition
      → The Ideology Machine
      → A Structural Theory Of Belief
      → Betrayal And Other Themes
      → Scripts And Screenplays
  3. Language And Understanding
    1. Responding To Language – 95
      → Programs That Pass As Persons
      → How Parry Manages It
      → How Eliza Differs From Eliza
    2. Intelligence In Understanding – 112
      → Conversation With Shrdlu
      → Heterarchical Thinking
      → What Shrdlu Knows
      → On Understanding Stories
    3. Sense And Semantics – 147
      → Knowing What's Going To Be Said
      → Programming Conceptual Dependencies
      → Machine Translation
      → Language Without Logic
  4. The Visual World
    1. Adding The Third Dimension – 179
      → Models, Cues, And Interpretation
      → How See Sees
      → Making Sense Of Pictures
    2. Glimpses Of The Real World –207
      → Hallucination In The Service Of Truth
      → Muffling The Combinatorial Explosion
      → Of Men And Movement
  5. New Thoughts From Old
    1. Learning – 247
      → Learning By Example
      → Learning By Being Told
      → Learning By Doing
    2. Creativity – 298
      → Creating Within A Convention
      → Fiction And Frames
      → Analogical Thinking
      → Changing Representations
    3. Problem Solving – 345
      → The Weakness Of Brute Force
      → Plans For Problems
      → Problems And Programming Languages
  6. The Relevance Of Artificial Intelligence
    1. Psychological Implications – 393
      → Humanism And Mechanism In Psychology
      → Theories And Testability
      → Machine Vision And Human Sight
    2. Philosophical Issues – 418
      → Artificial Intelligence: A Contradiction In Terms?
      → Mind, Mechanism, And Machines
      → Can Machine Intelligence Be Achieved?
    3. Social Significance – 445
      → Programs In Prospect
      → Selves And Society
      → Precautionary Measures
    4. Postscript – 474
    Bibliography – 501
    Bibliography To Postscript – 521
    Notes – 529
    Index – 567

In-Page Footnotes ("Boden (Margaret) - Artificial Intelligence and Natural Man")

Footnote 1:

The MIT Press, London, Second Edition (Expanded), 1987

Text Colour Conventions (see disclaimer)
  1. Blue: Text by me; © Theo Todman, 2021
  2. Mauve: Text by correspondent(s) or other author(s); © the author(s)

© Theo Todman, June 2007 - Jan 2021. Please address any comments on this page to File output:
Website Maintenance Dashboard
Return to Top of this Page Return to Theo Todman's Philosophy Page Return to Theo Todman's Home Page