Susan G. Sterrett

I argue here that Frege’s eventual view on the relation between sentences and the thoughts they express is that, ideally, a sentence expresses exactly one thought, and a thought is expressed by exactly one (canonical) sentence. This may clash with some mainstream views of Frege, for it has the consequence of de-emphasizing the philosophical significance of the question of how it is possible for someone to regard one sentence as true yet regard another sentence that expresses the same thought as false. This account of Frege was developed by taking a long-range look at his writings over the course of his life.

Abstract Philosophical discussion of Alan Turing’s writings on intelligence has mostly revolved around a single point made in a paper published in the journal Mind in 1950. This is unfortunate, for Turing’s reflections on machine (artificial) intelligence, human intelligence, and the relation between them were more extensive and sophisticated. They are seen to be extremely well-considered and sound in retrospect. Recently, IBM developed a question-answering computer (Watson) that could compete against humans on the game show Jeopardy! There are hopes it can be adapted to other contexts besides that game show, in the role of a collaborator of, rather than a competitor to, humans. Another, different, research project --- an artificial intelligence program put into operation in 2010 --- is the machine learning program NELL (Never Ending Language Learning), which continuously ‘learns’ by ‘reading’ massive amounts of material on millions of web pages. Both of these recent endeavors in artificial intelligence rely to some extent on the integration of human guidance and feedback at various points in the machine’s learning process. In this paper, I examine Turing’s remarks on the development of intelligence used in various kinds of search, in light of the experience gained to date on these projects.

The concept of similar systems arose in physics, and appears to have originated with Newton in the seventeenth century. This chapter provides a critical history of the concept of physically similar systems, the twentieth century concept into which it developed. The concept was used in the nineteenth century in various fields of engineering, theoretical physics and theoretical and experimental hydrodynamics. In 1914, it was articulated in terms of ideas developed in the eighteenth century and used in nineteenth century mathematics and mechanics: equations, functions and dimensional analysis. The terminology physically similar systems was proposed for this new characterization of similar systems by the physicist Edgar Buckingham. Related work by Vaschy, Bertrand, and Riabouchinsky had appeared by then. The concept is very powerful in studying physical phenomena both theoretically and experimentally. As it is not currently part of the core curricula of STEM disciplines or philosophy of science, it is not as well known as it ought to be.

Summary Analogue models are actual physical setups used to model something else. They are especially useful when what we wish to investigate is difficult to observe or experiment upon due to size or distance in space or time: for example, if the thing we wish to investigate is too large, too far away, takes place on a time scale that is too long, does not yet exist or has ceased to exist. The range and variety of analogue models is too extensive to attempt a survey. In this article, I describe and discuss several different analogue model experiments, the results of those model experiments, and the basis for constructing them and interpreting their results. Examples of analogue models for surface waves in lakes, for earthquakes and volcanoes in geophysics, and for black holes in general relativity, are described, with a focus on examining the bases for claims that these analogues are appropriate analogues of what they are used to investigate. A table showing three different kinds of bases for reasoning using analogue models is provided. Finally, it is shown how the examples in this article counter three common misconceptions about the use of analogue models in physics.

I address questions about values in model-making in engineering, specifically: Might the role of values be attributable solely to interests involved in specifying and using the model? Selected examples illustrate the surprisingly wide variety of things one must take into account in the model-making itself. The notions of system , and physically similar systems are important and powerful in determining what is relevant to an engineering model. Another example illustrates how an idea to completely re-characterize, or reframe, an engineering problem arose during model-making.I employ a qualitative analogue of the notion of physically similar systems. Historical cases can thus be drawn upon; I illustrate with a comparison between a geoengineering proposal to inject, or spray, sulfate aerosols, and two different historical cases involving the spraying of DDT . The current geoengineering proposal is seen to be like the disastrous and counterproductive case, and unlike the successful case, of the spraying of DDT. I conclude by explaining my view that model-making in science is analogous to moral perception in action, drawing on a view in moral theory that has come to be called moral particularism

On a literal reading of `Computing Machinery and Intelligence'', Alan Turing presented not one, but two, practical tests to replace the question `Can machines think?'' He presented them as equivalent. I show here that the first test described in that much-discussed paper is in fact not equivalent to the second one, which has since become known as `the Turing Test''. The two tests can yield different results; it is the first, neglected test that provides the more appropriate indication of intelligence. This is because the features of intelligence upon which it relies are resourcefulness and a critical attitude to one''s habitual responses; thus the test''s applicablity is not restricted to any particular species, nor does it presume any particular capacities. This is more appropriate because the question under consideration is what would count as machine intelligence. The first test realizes a possibility that philosophers have overlooked: a test that uses a human''s linguistic performance in setting an empirical test of intelligence, but does not make behavioral similarity to that performance the criterion of intelligence. Consequently, the first test is immune to many of the philosophical criticisms on the basis of which the (so-called) `Turing Test'' has been dismissed.

The topic of this Handbook entry is the relationship between similarity and dimensional analysis, and some of the philosophical issues involved in understanding and making use of that relationship. Discusses basics of the relationship between units, dimensions, and quantities. It explains the significance of dimensionless parameters, and explains that similarity of a physical systems is established by showing equality of a certain set of dimensionless parameters that characterizes the system behavior. Similarity is always relative -- to some system behavior. Other topics discussed: generalization of the notion of similarity, the difference between relative similarity and partial similarity; how the notion of similarity in science differs from similarity as it has been discussed in recent philosophy. Philosophers' views discussed: R. Giere, N. Goodman, P. Bridgman, and B. Ellis.

The year that Ludwig Wittgenstein was born in Vienna, 1889, nearby developments already underway portended two major changes of the coming century: the advent of controlled heavier-than-air flight and the mass production of musical sound recordings. Before they brought about major social changes, though, these innovations appeared in Europe in the form of children’s toys. Both a rubber-band-powered model helicopter-like toy employing an ingenious solution to the problem of control, and a working toy gramophone with which music could be reproduced from hard discs, appeared in Europe in time for Ludwig’s childhood. And, both innovations reappear in his work as an adult. The relationship between the advent of heavier-than-air flight and Wittgenstein’s claim in Tractatus Logico-Philosophicus that a proposition is a picture or model is a topic in its own right, and I discuss it in separate works. 1 In this essay, I consider the way Wittgenstein employed the development of sound recordings in discussing logical form in the Tractatus

Philosophical discussion of Alan Turing’s writings on intelligence has mostly revolved around a single point made in a paper published in the journal Mind in 1950. This is unfortunate, for Turing’s reflections on machine (artificial) intelligence, human intelligence, and the relation between them were more extensive and sophisticated. They are seen to be extremely well-considered and sound in retrospect. Recently, IBM developed a question-answering computer (Watson) that could compete against humans on the game show Jeopardy! There are hopes it can be adapted to other contexts besides that game show, in the role of a collaborator of, rather than a competitor to, humans. Another, different, research project—an artificial intelligence program put into operation in 2010—is the machine learning program NELL (Never Ending Language Learning), which continuously ‘learns’ by ‘reading’ massive amounts of material on millions of web pages. Both of these recent endeavors in artificial intelligence rely to some extent on the integration of human guidance and feedback at various points in the machine’s learning process. In this paper, I examine Turing’s remarks on the development of intelligence used in various kinds of search, in light of the experience gained to date on these projects.

Ernst Mach is the only person whom Einstein included on both the list of physicists he considered his true precursors, and the list of the philosophers who had most affected him. Einstein scholars have been less generous in their estimation of Mach's contributions to Einstein's work, and even amongst the more generous of them, Mach's great achievements in physics are seldom mentioned in this context. This is odd, considering Mach was nominated for the Nobel Prize in Physics three times. In this paper, I examine some of Mach's work in physics that bears conceptually on Einstein's 1905 paper on Special Relativity ("On The Electrodynamics of Moving Bodies"). Mach was the first to give the correct explanation of the Doppler Effect, and he presented it in a way that Einstein echoes in his 1905 paper: laying out two apparently contradictory principles and showing how both can be retained. It is also notable that Mach's explanation was explicit about not relying on the existence of a medium of transmission for the propagation of light waves. In his work on supersonic shock waves, Mach invokes the constancy of the velocity of sound (i.e., its independence of the motion of the sound source) , just as he had invoked the constancy of the velocity of light in his work on the Doppler Effect for Light. I examine the analogies between light and sound that were drawn upon by Einstein and Mach, as well as one analogy that Einstein could have, but did not make: Cherenkov radiation, or "singing electrons", i.e., cases in which the sound of light in the medium of transmission is exceeded, which results in an optical analogue of supersonic shock waves

Experimental engineering models have been used both to model general phenomena, such as the onset of turbulence in fluid flow, and to predict the performance of machines of particular size and configuration in particular contexts. Various sorts of knowledge are involved in the method - logical consistency, general scientific principles, laws of specific sciences, and experience. I critically examine three different accounts of the foundations of the method of experimental engineering models (scale models), and examine how theory, practice, and experience are involved in employing the method to obtain practical results. Models of machines and mechanisms can be (and generally are) involved in establishing criteria for similar phenomena, which provide guidance in using events to model other events. Conversely, models of phenomena such as events that model other events can be (and generally are) involved in experimentation on models of machines. I conclude that often it is not more detailed models or the more precise equations they engender that leads to better understanding, but rather an insightful use of knowledge at hand to determine which similarity principles are appropriate in allowing us to infer what we do not know from what we are able to observe.

In this paper I discuss the relationship between model, theories, and laws in the practice of experimental scale modeling. The methodology of experimental scale modeling, also known as physical similarity, differs markedly from that of other kinds of models in ways that are important to issues in philosophy of science. Scale models are not discussed in much depth in mainstream philosophy of science. In this paper, I examine how scale models are used in making inferences. The main question I address in this talk is ``How are fundamental laws involved in the construction of, and inferences drawn from, experimental scale models?'' We shall see that there is a refreshing alternative to the mainstream view that models can serve only as intermediaries between theory and experiment. Using the methodology of scale models, one can use observations on one piece of the world to make inferences about another piece of the world, without involving an intermediate abstract model about which one reasons. The philosophical significance of that point to philosophy of science is that the method of physical similarity, which provides the basis for inferences based upon scale models, is a qualitatively different way in which fundamental laws can be used in analogical reasoning that is truly informative. Finally, as this method provides a formal basis for case-based reasoning, it may be helpful in formalizing methods used in some of the so-called ``special sciences''.

Today I want to talk about an element in the milieu in which Ludwig Wittgenstein conceived the Tractatus Logico-Philosophicus that has not been recognized to date: the generalization of the methodology of experimental scale models that occurred just about the time he was writing it. I find it very helpful to keep in mind how this kind of model portrays when reading the Tractatus — in particular, when reading the statements about pictures and models, such as:That a picture is a fact ,That a picture is a model of reality ,That the “pictorial relationship” that makes a picture a picture is part of that picture , andThat a picture must have its pictorial form in common with reality in order to able to depict it

Turing wrote that the “guiding principle” of his investigation into the possibility of intelligent machinery was “The analogy [of machinery that might be made to show intelligent behavior] with the human brain.” [10] In his discussion of the investigations that Turing said were guided by this analogy, however, he employs a more far-reaching analogy: he eventually expands the analogy from the human brain out to “the human community as a whole.” Along the way, he takes note of an obvious fact in the bigger scheme of things regarding human intelligence: grownups were once children; this leads him to imagine what a machine analogue of childhood might be. In this paper, I’ll discuss Turing’s child-machine, what he said about different ways of educating it, and what impact the “bringing up” of a child-machine has on its ability to behave in ways that might be taken for intelligent. I’ll also discuss how some of the various games he suggested humans might play with machines are related to this approach.

I survey a broad variety of models with an eye to asking what kind of model each is in the following sense: in virtue of what is each of them regarded as a model? It will be seen that when we classify models according to the answer to this question, it comes to light that the notion of model predominant in philosophy of science covers only some of the kinds of models used in scientific contexts. The notion of a model predominant in philosophy of science requires that a model be related to something formal, such as equations or statements. Not all the examples provided in the brief survey in this paper fit that notion of a model. I identify another kind of model that ought to be included in philosophical and foundational studies of scientific models, which I call a “piece of the world” kind of model, to contrast with a “realm of thought” kind of model. These models also have formal methodologies associated with them, and, hence, analytic philosophers of science can embrace them without abandoning the rigor that has characterized the discipline.

How are beliefs efficacious? One answer is: via rational intentional action. But there are other ways that beliefs are efficacious. This dissertation examines these other ways, and sketches an answer to the question of how beliefs are efficacious that takes into account how beliefs are involved in the full range of behavioral disciplines, from psychophysiology and cognition to social and economic phenomena. The account of how beliefs are efficacious I propose draws on work on active accounts of perception. I develop an account based on a proposal sketched by the cognitive scientist Ulrich Neisser. Neisser sketched an active account of perception, on which dynamic anticipatory schemata direct an organism's exploration and action, and are in turn revised as a result of exploration and action. This notion of schema has roots in nineteenth century neurophysiology and in Frederick Bartlett's subsequent work on memory. Neisser appealed to it to unite what he thought was right about information-processing accounts of perception with what he thought was right about ecological accounts of perception. The point that we must anticipate in order to perceive has been recognized by philosophers in the form of the "theory-ladenness of observation." I extend the concept of anticipatory schema to include its role in social perception and social interaction; the concept of anticipatory schema provides a more interactive account of the role of expectations in the maintenance and existence of social institutions, and can be used to enrich the account of convention David Lewis provided. I also show that the concept of rational expectations, which explains the neutrality of money in terms of the efficacy of anticipatory expectations, is compatible with the proposed account of how beliefs are efficacious. I discuss how the proposal accounts for the three main modes by which beliefs can be efficacious: via their role in causing intentional action, via their role in causing economic phenomena and the existence and maintenance of social institutions, and via their role in causing unintentional physiological responses, including anticipatory physiological responses that can enable perception, cause involuntary actions and give rise to the placebo effect.

The analogy Darwin drew between artificial and natural selection in "On the Origin of Species" has a detailed structure that has not been appreciated. In Darwin’s analogy, the kind of artificial selection called Methodical selection is analogous to the principle of divergence in nature, and the kind of artificial selection called Unconscious selection is analogous to the principle of extinction in nature. This paper argues that it is the analogy between these two different principles familiar from his studies of artificial selection and the two different principles he claims are operative in nature that provides the main structure and force of the analogy he uses to make his case for the power of natural selection to produce new species. Darwin’s statements explicitly distinguishing between these two kinds of principles at work in nature occur prominently in the text of the Origin. The paper also shows that a recent revisionist claim that Darwin did not appeal to the efficacy of artificial selection is mistaken

. Some disagreements have arisen in the last few years regarding the role played by material properties when modeling, simulating and experimenting on physical systems (Morrison 2008, Parker (forthcoming), Winsberg (forthcoming), Guala 2002, 2005; Morgan 2005). The question has proven more involved than it first appears. A number of significant and correct points have already been made, but some confusions remain. In this paper I attempt to sort them out. After pointing out the importance of some distinctions that need to be drawn, I explain the ways in which material properties matter in modeling and simulations.

In "The Status and Future of the Turing Test" (Moor, 2001), which appeared in an earlier issue of this journal, James Moor remarks on my paper "Turing's Two Tests for Intelligence." In my paper I had claimed that, whatever Turing may or may not have thought, the test described in the opening section of Turing's now legendary 1950 paper "Computing Machinery and Intelligence" is not equivalent to, and in fact is superior to, the test described in a passage that occurs much later in Turing's paper (i.e., in Section 5 of Turing, 1950). I'm pleased Moor chose to give such prominence to my point, and very happy to see that he recognized that my claim was a normative one about the superiority of one test over another, rather than a claim about Turing's intentions. However, as I think the way he describes my point could lead to misunderstandings, I'd like to clarify the points I made. One major clari?cation is which two tests I am contrasting; another is that the difference in overall structure of the two tests is of philosophical signi?cance

Demonstrating the different roles that logic plays in the disciplines of computer science, mathematics, and philosophy, this concise undergraduate textbook covers select topics from three different areas of logic: proof theory, computability theory, and nonclassical logic. The book balances accessibility, breadth, and rigor, and is designed so that its materials will fit into a single semester. Its distinctive presentation of traditional logic material will enhance readers' capabilities and mathematical maturity. The proof theory portion presents classical propositional logic and first-order logic using a computer-oriented (resolution) formal system. Linear resolution and its connection to the programming language Prolog are also treated. The computability component offers a machine model and mathematical model for computation, proves the equivalence of the two approaches, and includes famous decision problems unsolvable by an algorithm. The section on nonclassical logic discusses the shortcomings of classical logic in its treatment of implication and an alternate approach that improves upon it: Anderson and Belnap's relevance logic. Applications are included in each section. The material on a four-valued semantics for relevance logic is presented in textbook form for the first time. Aimed at upper-level undergraduates of moderate analytical background, Three Views of Logic will be useful in a variety of classroom settings. Gives an exceptionally broad view of logic. Treats traditional logic in a modern format. Presents relevance logic with applications. Provides an ideal text for a variety of one-semester upper-level undergraduate courses

Twenty years ago in "Turing's Two Tests for Intelligence" I distinguished two distinct tests to be found in Alan Turing's 1950 paper "Computing Machinery and Intelligence": one by then very well-known, the other neglected. I also explained the significance of the neglected test. This paper revisits some of the points in that paper and explains why they are even more relevant today. It also discusses the value of tests for machine intelligence based on games humans play, giving an analysis of some twentieth century TV game shows and how they relate to the tests for machine intelligence in Turing's paper and in some other tests for machine intelligence that have been proposed since. Their value in distinguishing between 'wise' and simply ‘clever’ AI is discussed.

This paper investigates the following proposal about machine intelligence: that behaviour in which a habitual response that would have been inappropriate in a certain unfamiliar situation is overridden and replaced by a more appropriate response be considered evidence of intelligence. The proposal was made in an earlier paper (Sterrett 2000) and arose from an analysis of a neglected test for intelligence hinted at in Turing's legendary 'Computing Machinery and Intelligence'; it was also argued there that it was a more principled test of machine intelligence than straightforward comparisons with human behaviour. The present paper first summarizes the previous claim then looks at writings about intelligence, or the lack of it, in animals and machines by various writers (Descartes, Hume, Darwin and James). It is then shown that, despite their considerable differences regarding fundamental things such as what kinds of creatures are intelligent and the relationship between reason, instinct and behaviour, all of these writers would regard behaviour that meets the proposed criterion as evidence of intelligence. Finally, some recent work in employing logic and reinforcement learning in conjunction with 'behaviour-based' principles in the design of intelligent agents is described; the significance for the prospect of machine intelligence according to the proposed criterion is discussed

Putting Wittgenstein's writing into an historical context that includes scientific and technological developments as well as cultural and intellectual works can be helpful in understanding some of Wittgenstein's works. I focus on the Tractatus Logico-Philosophicus in particular in this paper, and on topics related to pictures and models: the development of audio recording technologies, the development of miniature scale models that were both aesthetically pleasing and scientifically useful, particularly in the forensics of traffic accidents, and the culmination of a centuries-long effort to articulate the method behind the use of physical modeling, i.e., the formulation of a concept presented in 1914 and dubbed "physically similar systems."