The application of statistical methods and models both in the natural and social sciences is nowadays a trivial fact which nobody would deny. Bold analogies even suggest the application of the same statistical models to fields as different as statistical mechanics and economics, among them the case of the young and controversial discipline of Econophysics . Less trivial, however, is the answer to the philosophical question, which has been raised ever since the possibility of “commuting” statistical (...) thinking and models between natural and social sciences emerged: whether such a methodological kinship would imply some kind of more profound unity of the natural and the social domain. Starting with Adolphe Quetelet (1796-1874) and ending with the Vienna Circle (from the late 1920s until the 1940s), this paper offers a brief historical and philosophical reconstruction of some important stages in the development of statistics as “commuting” between the natural and the social sciences. This reconstruction is meant to highlight (with respect to the authors under consideration): (1) the existence of a significant correlation between the readiness to “transfer” statistical thinking from natural to social sciences and vice versa, on the one hand, and the standpoints on the issue of the unity/disunity of science, on the other; (2) the historical roots and the fortunes of the analogy between statistical models of society and statistical models of gases. (shrink)

What Chow calls NHSTP is an inconsistent hybrid of Fisherian and Neyman-Pearsonian ideas. In psychology it has been practiced like ritualistic handwashing and sustained by wishful thinking about its utility. Chow argues that NHSTP is an important tool for ruling out chance as an explanation for data. I disagree. This ritual discourages theory development by providing researchers with no incentive to specify hypotheses.

While correctly describing the differences in current practices between experimental psychologists and economists, Hertwig and Ortmann do not provide a compelling explanation for these differences. Our explanation focuses on the fact that psychologists view the world as composed of categories and types. This discrete organizational scheme results in merely testing nulls and wider variation in observed practices in experimental psychology.

The law views with suspicion statistical evidence, even evidence that is probabilistically on a par with direct, individual evidence that the law is in no way suspicious of. But it has proved remarkably hard to either justify this suspicion, or to debunk it. In this paper, we connect the discussion of statistical evidence to broader epistemological discussions of similar phenomena. We highlight Sensitivity – the requirement that a belief be counterfactually sensitive to the truth in a specific way (...) – as a way of epistemically explaining the legal suspicion towards statistical evidence. Still, we do not think of this as a satisfactory vindication of the reluctance to rely on statistical evidence. Knowledge – and Sensitivity, and indeed epistemology in general – are of little, if any, legal value. Instead, we tell an incentive-based story vindicating the suspicion towards statistical evidence. We conclude by showing that the epistemological story and the incentive-based story are closely and interestingly related, and by offering initial thoughts about the role of statistical evidence in morality. (shrink)

Winner of the National Business Book Award From the New York Times bestselling author of The Organized Mind and This Is Your Brain on Music, a primer to the critical thinking that is more necessary now than ever We are bombarded with more information each day than our brains can process—especially in election season. It's raining bad data, half-truths, and even outright lies. New York Times bestselling author Daniel J. Levitin shows how to recognize misleading announcements, statistics, graphs, and (...) written reports, revealing the ways lying weasels can use them. It's becoming harder to separate the wheat from the digital chaff. How do we distinguish misinformation, pseudo-facts, and distortions from reliable information? Levitin groups his field guide into two categories—statistical information and faulty arguments—ultimately showing how science is the bedrock of critical thinking. Infoliteracy means understanding that there are hierarchies of source quality and bias that variously distort our information feeds via every media channel, including social media. We may expect newspapers, bloggers, the government, and Wikipedia to be factually and logically correct, but they so often aren't. We need to think critically about the words and numbers we encounter if we want to be successful at work, at play, and in making the most of our lives. This means checking the plausibility and reasoning—not passively accepting information, repeating it, and making decisions based on it. Readers learn to avoid the extremes of passive gullibility and cynical rejection. Levitin's charming, entertaining, accessible guide can help anyone wake up to a whole lot of things that aren't so. And catch some weasels in their tracks! (shrink)

Recently, Jennifer Saul (“Racial Figleaves, the Shifting Boundaries of the Permissible, and the Rise of Donald Trump”, 2017; “Racist and Sexist Figleaves”, 2021) has explored the use of what she calls “figleaves” in the discourse on race and gender. Following Saul, a figleaf is an utterance that, for some portion of the audience, blocks the conclusion that some other utterance, R, or the person who uttered R is racist or sexist. Such racial and gender figleaves are pernicious, says Saul, because, (...) among other things, they can shift the boundaries of what is deemed acceptable to think or say. This paper expands on Saul’s picture in a threefold way: It is first argued that appeals to statistics can function as figleaves. It is then argued that—as far as figleaves go—there is reason to believe that statistics-as-figleaves are especially pernicious. Finally, the paper explores some strategies for counterspeech. (shrink)

After some general remarks about the interrelation between philosophical and statistical thinking, the discussion centres largely on significance tests. These are defined as the calculation of p-values rather than as formal procedures for ‘acceptance‘ and ‘rejection‘. A number of types of null hypothesis are described and a principle for evidential interpretation set out governing the implications of p- values in the specific circumstances of each application, as contrasted with a long-run interpretation. A number of more complicated situ- ations (...) are discussed in which modification of the simple p-value may be essential. (shrink)

Albert Einstein's attitude towards quantum mechanics—and statistical physics in general—was a puzzle to many of his contemporaries, and has remained a puzzle to the present. Though he made many significant contributions to statistical physics, he continually refused to regard that branch of science as fundamental. The present essay demonstrates that his attitude towards statistical physics was formed during his earliest investigations—between 1901 and 1903. In particular, it is shown that in Einstein's view, statistical laws are based (...) upon non-statistical assumptions. This view influenced much of his later thinking. A few illustrations are offered, concluding with a discussion of his opposition to quantum mechanics. (shrink)

In the late 19th century, the questionnaire was one means of taking the case study into the multitudes. This article engages with Forrester’s idea of thinking in cases as a means of interrogating questionnaire-based research in early American psychology. Questionnaire research was explicitly framed by psychologists as a practice involving both natural historical and statistical forms of scientific reasoning. At the same time, questionnaire projects failed to successfully enact the latter aspiration in terms of synthesizing masses of collected (...) data into a coherent whole. Difficulties in managing the scores of descriptive information questionnaires generated ensured the continuing presence of individuals in the results of this research, as the individual case was excerpted and discussed alongside a cast of others. As a consequence, questionnaire research embodied an amalgam of case, natural historical, and statistical thinking. Ultimately, large-scale data collection undertaken with questionnaires failed in its aim to construct composite exemplars or ‘types’ of particular kinds of individuals; to produce the singular from the multitudes. (shrink)

Achieving a good clinical trial design increases the likelihood that a trial will take place as planned, including that data will be obtained from a sufficient number of participants, and the total number of participants will be the minimal required to gain the knowledge sought. A good trial design also increases the likelihood that the knowledge sought by the experiment will be forthcoming. Achieving such a design is more than good sense—it is ethically required in experiments when participants are at (...) risk of harm. This paper argues that doing a power analysis effectively contributes to ensuring that a trial design is good. The ethical importance of good trial design has long been recognized for trials in which there is risk of serious harm to participants. However, whether the quality of a trial design, when the risk to participants is only minimal, is an ethical issue is rarely discussed. This paper argues that even in cases when the risk is minimal, the quality of the trial design is an ethical issue, and that this is reflected in the emphasis the Belmont Report places on the importance of the benefit of knowledge gained by society. The paper also argues that good trial design is required for true informed consent. (shrink)

This study is intended to provide in-depth insights into how design thinking and creativity issues are understood and possibly evolve in the course of design discussions in a company context. For that purpose, we use the seminar transcripts of the Design Thinking Research Symposium 12 (DTRS12) dataset “Tech-centred Design Thinking: Perspectives from a Rising Asia,” which are primarily concerned with how Korean companies implement design thinking and what role designers currently play. We employed a novel method (...) of information processing based on constructed dynamic semantic networks to investigate the seminar discussions according to company representatives and company size. We compared the quantitative dynamics in two seminars: the first involved managerial representatives of four companies, and the second involved specialized designers and management of a design center of single company. On the basis of dynamic semantic networks, we quantified the changes in four semantic measures—abstraction, polysemy, information content, and pairwise word similarity—in chronologically reconstructed individual design-thinking processes. Statistical analyses show that design thinking in the seminar with four companies, exhibits significant differences in the dynamics of abstraction, polysemy, and information content, compared to the seminar with the design center of single company. Both the decrease in polysemy and abstraction and the increase in information content in the individual design-thinking processes in the seminar with four companies indicate that design managers are focused on more concrete design issues, with more information and less ambiguous content to the final design product. By contrast, specialized designers manifest more abstract thinking and appear to exhibit a slightly higher level of divergence in their design processes. The results suggest that design thinking and creativity issues are articulated differently depending on designer roles and the company size. (shrink)

Statistical Learning Theory (e.g., Hastie et al., 2001; Vapnik, 1998, 2000, 2006) is the basic theory behind contemporary machine learning and data-mining. We suggest that the theory provides an excellent framework for philosophical thinking about inductive inference.

Statistical tests of the primality of some numbers look similar to statistical tests of many nonmathematical, clearly empirical propositions. Yet interpretations of probability prima facie appear to preclude the possibility of statistical tests of mathematical propositions. For example, it is hard to understand how the statement that n is prime could have a frequentist probability other than 0 or 1. On the other hand, subjectivist approaches appear to be saddled with ‘coherence’ constraints on rational probabilities that require (...) rational agents to assign extremal probabilities to logical and mathematical propositions. In the light of these problems, many philosophers have come to think that there must be some way to generalize a Bayesian statistical account. In this article I propose that a classical frequentist approach should be reconsidered. I conclude that we can give a conditional justification of statistical testing of at least some mathematical hypotheses: if statistical tests provide us with reasons to believe or bet on empirical hypotheses in the standard situations, then they also provide us with reasons to believe or bet on mathematical hypotheses in the structurally similar mathematical cases. (shrink)

Critical thinking and arguments -- What makes a good argument? -- Premises and conclusions -- Language -- Propositional arguments -- Categorical arguments -- Analogical arguments -- Statistical arguments -- Causal arguments -- Moral arguments -- Answers to selected exercises -- Reference guide.

Many who think that naked statistical evidence alone is inadequate for a trial verdict think that use of probability is the problem, and something other than probability – knowledge, full belief, causal relations – is the solution. I argue that the issue of whether naked statistical evidence is weak can be formulated within the probabilistic idiom, as the question whether likelihoods or only posterior probabilities should be taken into account in our judgment of a case. This question also (...) identifies a major difference between the Process Reliabilist and Probabilistic Tracking views of knowledge and other concepts. Though both are externalist, and probabilistic, epistemic theories, Tracking does and Process Reliabilism does not put conditions on likelihoods. So Tracking implies that a naked statistic is not adequate evidence about an individual, and does not yield knowledge, whereas the Reliabilist thinks it gives justified belief and knowledge. Not only does the Tracking view imply that naked statistical evidence is insufficient for a verdict, but it gives us resources to explain why, in terms of risk and the special conditions of a trial. (shrink)

As artificial intelligence becomes ubiquitous, it will be increasingly involved in novel, morally significant situations. Thus, understanding what it means for a machine to be morally responsible is important for machine ethics. Any method for ascribing moral responsibility to AI must be intelligible and intuitive to the humans who interact with it. We argue that the appropriate approach is to determine how AIs might fare on a standard account of human moral responsibility: a Strawsonian account. We make no claim that (...) our Strawsonian approach is either the only one worthy of consideration or the obviously correct approach, but we think it is preferable to trying to marry fundamentally different ideas of moral responsibility into a single cohesive account. Under a Strawsonian framework, people are morally responsible when they are appropriately subject to a particular set of attitudes—reactive attitudes—and determine under what conditions it might be appropriate to subject machines to this same set of attitudes. Although the Strawsonian account traditionally applies to individual humans, it is plausible that entities that are not individual humans but possess these attitudes are candidates for moral responsibility under a Strawsonian framework. We conclude that weak AI is never morally responsible, while a strong AI with the right emotional capacities may be morally responsible. (shrink)

In this paper I critically examine the Generality Problem and argue that it does not succeed as an objection to reliabilism. Although those who urge the Generality Problem are correct in claiming that any process token can be given indefinitely many descriptions that pick out indefinitely many process types, they are mistaken in thinking that reliabilists have no principled way to distinguish between relevant and irrelevant process types.

This textbook carefully develops the main ideas and techniques of statistical and thermal physics and is intended for upper-level undergraduate courses. The authors each have more than thirty years' experience in teaching, curriculum development, and research in statistical and computational physics. Statistical and Thermal Physics begins with a qualitative discussion of the relation between the macroscopic and microscopic worlds and incorporates computer simulations throughout the book to provide concrete examples of important conceptual ideas. Unlike many contemporary texts (...) on thermal physics, this book presents thermodynamic reasoning as an independent way of thinking about macroscopic systems. Probability concepts and techniques are introduced, including topics that are useful for understanding how probability and statistics are used. Magnetism and the Ising model are considered in greater depth than in most undergraduate texts, and ideal quantum gases are treated within a uniform framework. Advanced chapters on fluids and critical phenomena are appropriate for motivated undergraduates and beginning graduate students. Integrates Monte Carlo and molecular dynamics simulations as well as other numerical techniques throughout the text Provides self-contained introductions to thermodynamics and statistical mechanics Discusses probability concepts and methods in detail Contains ideas and methods from contemporary research Includes advanced chapters that provide a natural bridge to graduate study Features more than 400 problems Programs are open source and available in an executable cross-platform format Solutions manual. (shrink)

The processes underwriting the acquisition of culture remain unclear. How are shared habits, norms, and expectations learned and maintained with precision and reliability across large-scale sociocultural ensembles? Is there a unifying account of the mechanisms involved in the acquisition of culture? Notions such as “shared expectations,” the “selective patterning of attention and behaviour,” “cultural evolution,” “cultural inheritance,” and “implicit learning” are the main candidates to underpin a unifying account of cognition and the acquisition of culture; however, their interactions require greater (...) specification and clarification. In this article, we integrate these candidates using the variational (free-energy) approach to human cognition and culture in theoretical neuroscience. We describe the construction by humans of social niches that afford epistemic resources called cultural affordances. We argue that human agents learn the shared habits, norms, and expectations of their culture through immersive participation in patterned cultural practices that selectively pattern attention and behaviour. We call this process “thinking through other minds” (TTOM) – in effect, the process of inferring other agents’ expectations about the world and how to behave in social context. We argue that for humans, information from and about other people's expectations constitutes the primary domain of statistical regularities that humans leverage to predict and organize behaviour. The integrative model we offer has implications that can advance theories of cognition, enculturation, adaptation, and psychopathology. Crucially, this formal (variational) treatment seeks to resolve key debates in current cognitive science, such as the distinction between internalist and externalist accounts of theory of mind abilities and the more fundamental distinction between dynamical and representational accounts of enactivism. (shrink)

When is a society egalitarian? When is it a meritocracy? The answers to these questions are not as obvious as some seem to think.

There are many objections to statistical discrimination in general and racial profiling in particular. One objection appeals to the idea that people have a right to be treated as individuals. Statistical discrimination violates this right because, presumably, it involves treating people simply on the basis of statistical facts about groups to which they belong while ignoring non-statistical evidence about them. While there is something to this objection—there are objectionable ways of treating others that seem aptly described (...) as failing to treat them as individuals—it needs to be articulated carefully. First, most people accept that many forms of statistical discrimination are morally unproblematic, let alone morally justified all things considered. Second, even treating people on the basis of putative non-statistical evidence relies on generalizations. Once we construe treating someone as an individual in a way that respects this fact, it becomes apparent: (1) that statistical discrimination is compatible with treating people as individuals, and (2) that one may fail to treat people as individuals even without engaging in statistical discrimination. Finally, there are situations involving the expression of messages of inclusion where we think it is good, morally speaking, that we are not treated as individuals. (shrink)

It is possible to pose many difficult and fascinating problems and criticisms for the various theses and arguments in Escape from Leviathan (EfL). This occurred while writing it, and various sharp minds did it on reading drafts or the final product. However, some reviews misunderstand, or ignore, what is written and reassert conventional views. But it is best to answer all published criticisms if only to show how they fail, lest anyone thinks they are sound, and even poor criticisms can (...) sometimes elicit elucidation. Thus, we now turn to the review. (shrink)

This paper is a discussion of David Albert's approach to the foundations of classical statistical menchanics. I point out a respect in which his account makes a stronger claim about the statistical mechanical probabilities than is usually made, and I suggest what might be motivation for this. I outline a less radical approach, which I attribute to Boltzmann, and I give some reasons for thinking that this approach is all we need, and also the most we are (...) likely to get. The issue between the two accounts turns out to be one about the explanatory role probabilities play in statistical mechanics. (shrink)

This essay proposes a design perspective on argumentation, intended as complementary to empirical and critical scholarship. In any substantive domain, design can provide insights that differ from those provided by scientific or humanistic perspectives. For argumentation, the key advantage of a design perspective is the recognition that humanity’s natural capacity for reason and reasonableness can be extended through inventions that improve on unaided human intellect. Historically, these inventions have fallen into three broad classes: logical systems, scientific methods, and disputation frameworks. (...) Behind each such invention is a specifiable “design hypothesis”: an idea about how to decrease error or how to increase the quality of outcomes from reasoning. As problems in contemporary argumentation practice become more complex, design thinking rises in relevance and importance. A design research agenda in argumentation would focus on theorizing design innovations and on evaluating design hypotheses. (shrink)

The second edition of a unique introductory text, offering an account of the logical tradition in philosophy and its influence on contemporary scientific disciplines. Thinking Things Through offers a broad, historical, and rigorous introduction to the logical tradition in philosophy and its contemporary significance. It is unique among introductory philosophy texts in that it considers both the historical development and modern fruition of a few central questions. It traces the influence of philosophical ideas and arguments on modern logic, statistics, (...) decision theory, computer science, cognitive science, and public policy. The text offers an account of the history of speculation and argument, and the development of theories of deductive and probabilistic reasoning. It considers whether and how new knowledge of the world is possible at all, investigates rational decision making and causality, explores the nature of mind, and considers ethical theories. Suggestions for reading, both historical and contemporary, accompany most chapters. This second edition includes four new chapters, on decision theory and causal relations, moral and political theories, “moral tools” such as game theory and voting theory, and ethical theories and their relation to real-world issues. Examples have been updated throughout, and some new material has been added. It is suitable for use in advanced undergraduate and beginning graduate classes in philosophy, and as an ancillary text for students in computer science and the natural sciences. (shrink)

How do fitness and natural selection relate to other evolutionary factors like architectural constraint, mode of reproduction, and drift? In one way of thinking, drawn from Newtonian dynamics, fitness is one force driving evolutionary change and added to other factors. In another, drawn from statistical thermodynamics, it is a statistical trend that manifests itself in natural selection histories. It is argued that the first model is incoherent, the second appropriate; a hierarchical realization model is proposed as a (...) basis for a statistical treatment. It emerges that natural selection does not cause evolution; it just is evolution. The theory incorporates relations of statistical correlation, but not the kind of causation found in fundamental physical processes. (shrink)

Statistics education at all levels includes data collected on human subjects. Thus, statistics educators have a responsibility to educate their students about the ethical aspects related to the collection of those data. The changing statistics education landscape has seen instruction moving from being formula-based to being focused on statistical reasoning. The widely implemented Guidelines for Assessment and Instruction in Statistics Education (GAISE) Report has paved the way for instructors to present introductory statistics to students in a way that is (...) both approachable and engaging. However, with technological advancement and the increase in availability of real-world datasets, it is necessary that instruction also integrate the ethical aspects around data sources, such as privacy, how the data were obtained and whether participants consent to the use of their data. In this article, we propose incorporating ethics into established curricula and integrating ethics into undergraduate-level introductory statistics courses based on recommendations in the GAISE Report. We provide a few examples of how to prompt students to constructively think about their ethical responsibilities when working with data. (shrink)

We are inundated by scientific and statistical information, but what should we believe? How much should we trust the polls on the latest electoral campaign? When a physician tells us that a diagnosis of cancer is 90% certain or a scientist informs us that recent studies support global warming, what should we conclude? How can we acquire reliable statistical information? Once we have it, how do we evaluate it? Despite the importance of these questions to our lives, many (...) of us have only a vague idea of how to answer them. In this admirably clear and engaging book, Mark Battersby provides a practical guide to thinking critically about scientific and statistical information. The goal of the book is not only to explain how to identify misleading statistical information, but also to give readers the understanding necessary to evaluate and use statistical and statistically based scientific information in their own decision making. (shrink)

Standard statistical measures of strength of association, although pioneered by Pearson deliberately to be acausal, nowadays are routinely used to measure causal efficacy. But their acausal origins have left them ill suited to this latter purpose. I distinguish between two different conceptions of causal efficacy, and argue that: 1) Both conceptions can be useful 2) The statistical measures only attempt to capture the first of them 3) They are not fully successful even at this 4) An alternative definition (...) more squarely based on causal thinking not only captures the second conception, it can also capture the first one better too. (shrink)

The Thinking Critically series introduces students to the complex issues that dominate public discourse and challenges them to become discerning readers, to think independently, and to engage and develop their skills as critical thinkers. Chapters are organized in a pro/con format, in which a single author synthesizes predominant arguments for and against an issue into clear, accessible discussions supported by details and evidence including relevant facts, direct quotes, current examples, and statistical illustrations. All volumes include focus questions to (...) guide students as they read each pro/con discussion, a list of key facts, source notes, and an annotated list of related organizations and websites for conducting further research. Book jacket. (shrink)

This book analyzes the origins of statistical thinking as well as its related philosophical questions, such as causality, determinism or chance. Bayesian and frequentist approaches are subjected to a historical, cognitive and epistemological analysis, making it possible to not only compare the two competing theories, but to also find a potential solution. The work pursues a naturalistic approach, proceeding from the existence of numerosity in natural environments to the existence of contemporary formulas and methodologies to heuristic pragmatism, a (...) concept introduced in the book's final section. This monograph will be of interest to philosophers and historians of science and students in related fields. Despite the mathematical nature of the topic, no statistical background is required, making the book a valuable read for anyone interested in the history of statistics and human cognition. (shrink)

Combining information from two or possibly several blocks of data is gaining increased attention and importance in several areas of science and industry. Typical examples can be found in chemistry, spectroscopy, metabolomics, genomics, systems biology and sensory science. Many methods and procedures have been proposed and used in practice. The area goes under different names: data integration, data fusion, multiblock analyses, multiset analyses and a few more. This book is an attempt to give an up-to-date treatment of the most used (...) and important methods within an important branch of the area; namely methods based on so-called components or latent variables. These methods have already obtained an enormous attention in for instance chemometrics, bioinformatics, machine learning and sensometrics and have proved to be important both for prediction and interpretation. The book is primarily a description of methodologies, but most of the methods will be illustrated by examples from the above-mentioned areas. The book is written such that both users of the methods as well as method developers will hopefully find sections of interest. In the end of the book there is a description of a software package developed particularly for the book. This package is freely available in R and covers many of the methods discussed. To distinguish the different type of methods from each other, the book is divided into five parts. Part I is introduction and preliminary concepts. Part II is the core of the book containing the main unsupervised and supervised methods. Part III deals with more complex structures and, finally, Part IV discusses alternative unsupervised and supervised methods. The book ends with Part V discussing the available software. Our recommendation for reading the book are as follows. A minimum read of the book would involve chapters 1, 2, 3, 5 and 7. Chapters 4, 6 and 8 are more specialized and chapters 9 and 10 contain methods we think are more advanced or less obvious to use. We feel privileged to have so many friendly colleagues who were willing to spend their time on helping us to improve the book by reading separate chapters. We would like to express our thanks to: Rasmus Bro, Margriet Hendriks, Ulf Indahl, Henk Kiers, Ingrid MaÌ⁽ge, Federico Marini, AÌ⁽smund Rinnan, Rosaria Romano, Lars Erik Solberg, Marieke Timmerman, Oliver Tomic, Johan Westerhuis and Barry Wise. Of course, the correctness of the final text is fully our responsibility! (shrink)

Good scientific research depends on critical thinking at least as much as factual knowledge; psychology is no exception to this rule. And yet, despite the importance of critical thinking, psychology students are rarely taught how to think critically about the theories, methods, and concepts they must use. This book shows students and researchers how to think critically about key topics such as experimental research, statistical inference, case studies, logical fallacies, and ethical judgments. Using updated research findings and (...) new insights, this volume provides a comprehensive overview of what critical thinking is and how to teach it in psychology. Written by leading experts in critical thinking in psychology, each chapter contains useful pedagogical features, such as critical-thinking questions, brief summaries, and definitions of key terms. It also supplies descriptions of each chapter author's critical-thinking experience, which evidences how critical thinking has made a difference to facilitating career development. (shrink)

Ernst Mayr has argued that Darwinian theory discredited essentialist modes of thought and replaced them with what he has called "population thinking". In this paper, I characterize essentialism as embodying a certain conception of how variation in nature is to be explained, and show how this conception was undermined by evolutionary theory. The Darwinian doctrine of evolutionary gradualism makes it impossible to say exactly where one species ends and another begins; such line-drawing problems are often taken to be the (...) decisive reason for thinking that essentialism is untenable. However, according to the view of essentialism I suggest, this familiar objection is not fatal to essentialism. It is rather the essentialist's use of what I call the natural state model for explaining variation which clashes with evolutionary theory. This model implemented the essentialist's requirement that properties of populations be defined in terms of properties of member organisms. Requiring such constituent definitions is reductionistic in spirit; additionally, evolutionary theory shows that such definitions are not available, and, moreover, that they are not needed to legitimize population-level concepts. Population thinking involves the thesis that population concepts may be legitimized by showing their connections with each other, even when they are not reducible to concepts applying at lower levels of organization. In the paper, I develop these points by describing Aristotle's ideas on the origins of biological variation; they are a classic formulation of the natural state model. I also describe how the development of statistical ideas in the 19th century involved an abandoning of the natural state model. (shrink)