The Origins of Positivism: The Contributions of Comte and Spencer

Jonathan H Turner. Handbook of Social Theory. Editor: George Ritzer & Barry Smart. 2001. Sage Publication.

By the beginning of the nineteenth century, the social world was increasingly viewed by Enlightenment thinkers as part of the natural universe; and indeed, many were coming to the conclusion that the natural and social sciences could be used to promote human progress. This perspective was not accepted by all; and in fact, it took well over a century for the various social sciences to become institutionalized inside and outside of academia. Still, for well over a century, beginning in the early 1700s, the idea that human beings and their social world could be studied scientifically had been gaining momentum; and by the time that Auguste Comte began to publish his Course of Positive Philosophy (1830-42) and to proclaim that the day of sociology had arrived, this was no longer such a radical idea.

The Origins of Comte’s Positivism

The young Comte was, like most scholars of his time on the European continent, a child of the Enlightenment, especially the Scientific Revolution which had begun to offer the hope that science could be used in the name of human progress. By the time that Comte had begun to write, the moral fervor of the French Philosophers had been combined and tempered with the view that science could be the tool for reconstructing society along more humane and just lines. While Sir Francis Bacon (1561-1626) had been the first figure in the Enlightenment to give articulate expression to the modern scientific method, legitimating the great achievements in astronomy during the sixteenth and seventeenth centuries, it was Isaac Newton’s law of gravity that provided a vision of what scientific inquiry could be: formal laws stating the fundamental relationships among basic properties of the universe. It would take over a century to see clearly that the discovery of such laws could better the conditions of humankind, but Newton provided the model of how elegant science could be. Comte would take the slowly accumulating recognition of science as the means for human progress and forge this recognition into sociological positivism.

The first clear evidence of the transition to seeing science as the key to reconstructing society can be found in the works of Charles Montesquieu (1689-1755), who engaged in analysis that suggested the possibilities for a science of society resembling Newton’s great law. In his The Spirit of the Laws (1748), Montesquieu advocated that society must be considered a ‘thing’; and as such, its fundamental properties and dynamics could be discovered through systematic observation and analysis. Many of the ideas in Comte’s synthesis in the next century—the search for laws, the hierarchy of the sciences, the movement of societies through stages, for example—are to be found in rudimentary form in Montesquieu. Later thinkers, particularly Jacques Turgot (1727-81) and Jean Condorcet (1743-94), further instilled in Comte the idea of human progress through stages, especially the movement of systems of ideas. These thinkers also codified the French Philosophers’ notions of social justice and societal betterment into a more scientific form of expression. Thus, a science of society was becoming not only possible, but in true Enlightenment fashion, it was to be used to construct a better society and, thereby, further human progress.

After leaving the Ecole Polytechnique, Comte began his collaboration with Claude-Henri de Saint-Simon (1760-1825). It was during this somewhat tumultuous collaboration, first as Saint-Simon’s secretary later as a junior peer, that most of the ideas that were to appear in The Course of Positive Philosophy took definite form. Comte felt that Saint-Simon and his ardent followers, known as the ‘Saint-Simonians,’ were too prone to ameliorative efforts without proper scientific understanding of the dynamics of human social organization. Still, Saint-Simon’s works provided the foundation for Comte’s Positive Philosophy. For, it was Saint-Simon who used the term ‘positive’ science to describe a study of humankind and society based upon empirical observations; it was Saint-Simon who had revitalized the organismic analogy, seeing society as a kind of organism whose laws of development and organization could be discovered and whose pathologies could be treated like those of a biological organism; it was Saint-Simon who postulated a law of history moving from a religious to positivistic basis; it was Saint-Simon who understood that positivism penetrated the sciences at different rates, first into physics and chemistry and later into physiology (including both biological and sociocultural organisms); and it was Saint-Simon who advocated a ‘terrestrial morality’ based upon a positivistic view of using observations to develop, test and implement the laws of human organization. Just where Saint-Simon’s work leaves off and Comte’s begins is not clear, but there can be no doubt that Comte took much from his mentor before their irrevocable break in 1824—a break that left Comte an intellectual isolate at the very time he was beginning to write The Course of Positive Philosophy. Indeed, Comte was reduced to menial teaching and tutoring jobs that were considerably beneath his intellect; and so, he was to pay a very high price for his abrasive personality.

In desperation, Comte proposed a series of public lectures to recapture his fading esteem; and even though several dozen eminent scientists subscribed to the lectures, he gave only three before the pressure of the enterprise proved too much and made him ill. Even when the lectures were revived later, some of the early subscribers once again appeared but were soon driven away by Comte’s personality. Thus, as Comte was writing the first serialized instalments of Positive Philosophy, he was becoming an intellectual outcast; and even when the first volume of this work appeared to critical acclaim, his ideas did not attract wide attention and his acclaim was short-lived. He had alienated almost everyone and had become the enemy of the Saint-Simonians. His marriage and friends began to fail him, and by the time that the last volume of Positive Philosophy was published, not a single review of it appeared in the French Press. The founder of positivism and sociology was, therefore, to be a failure in his host country, although British social philosophers like John Stuart Mill and, most importantly, Herbert Spencer had read Comte with great interest. Thus, positivism was created and given its most articulate expression by a failing scholar, one whose star had fallen and one who would later become a pathetic figure proclaiming himself the Great Priest of Humanity and preaching to rag-tag groups of followers.

The Positive Philosophy of Auguste Comte

In 1822, Auguste Comte published the first clear statement of his positive philosophy in an article titled ‘Plan of the Scientific Operations Necessary for Reorganizing Society.’ For Comte, it was essential to create a ‘positive science’ like other sciences, and this science would be based upon empirical observations that would be used to generate and test abstract laws of human organization. This new science was to be called ‘social physics’ and once the laws of human organization have been discovered and formulated, they should be used to direct the operation of society. Scientists of society were, therefore, to guide the course and direction of human organization. One of the most fundamental laws of human organization was the ‘law of the three stages’—an idea which he clearly borrowed from Turgot, Condorcet and Saint-Simon—with each stage being typified by a particular kind of ‘spirit’—a notion that first appeared in Montesquieu’s The Spirit and was reinforced by Condorcet and Saint-Simon. These well-known stages were the ‘theological-military,’ ‘metaphysical-judicial,’ and ‘scientific industrial’ or positivistic. Society had now entered the last stage, and hence, it was possible to have a true science of social organization. For Comte, the age of sociology had arrived; and it was to be very much like Newtonian physics in the formulation of abstract laws on the forces of the social universe that could then be used to reconstruct society. This first important essay, written independently of Saint-Simon, presented in broad strokes the outline of Comte’s more ambitious Course of Positive Philosophy.

The Course of Positive Philosophy is a long work; and its goal was to unify all of the sciences, while advocating a place for sociology among the sciences. In many ways, Positive Philosophy is a history of science through the prism of the law of three stages and an effort to establish a program for the new science of society with respect to (1) theory, (2) methods, (3) substance and (4) advocacy. Each of these points of emphasis will guide the review of how Comte formulated his positivism.

The Nature of Sociological Theory

The opening pages of Positive Philosophy are filled with statements about the nature of theory in the new science. Comte was, like all thinkers of the Enlightenment, impressed with Newton’s law of gravity, and he felt that sociology could develop similar laws. As he emphasized (Comte, 1854: 5-6).

The first characteristic of Positive Philosophy is that it regards all phenomena as subject to invariable natural laws. Our business is—seeing how vain is any research into what are called Causeswhether first or final—to pursue an accurate discovery of these Laws, with a view to reducing them to the smallest possible number. By speculating upon causes, we could solve no difficulty about origins and purpose. Our real business is to analyse accurately the circumstances of phenomena, and to connect them by the natural relations of succession and resemblance. The best illustration of this is in the case of the doctrine of Gravitation.

This short quotation introduces a number of important issues which are critical to the positivist project as it was to unfold over the next one hundred and fifty years. First, there is the obvious reference to the goal of all theory: to articulate abstract laws about the operation of the social universe.

Second, is the nature of these laws, but here matters become a bit vague. In the context of Comte’s time, the reference to ‘first causes’ had several meanings: (a) first in the sense of God, (h) first in the sense of what initiated a phenomenon in the distant past and (c) first in the sense of the more proximate forces that set a phenomenon in motion. Search for such causes creates problems of placing trust in non-worldly entities (that is, God), of seeking the ‘Big Bang’ of a social phenomenon in the ultimate past, and of engaging in an infinite causal regress (that is, if A is caused by B, what caused B? Perhaps C, which was caused by D, and so on in a constant regress). In place of a concern with causality, Comte offered the notion of ‘natural relations of succession and resemblance.’ But, what does this mean? If the Law of Gravitation is the ideal, then it must mean that a law states relations among basic forces, as is the case where the magnitude of gravitation is postulated to be a function (multiplied by a constant) of the relative size of the respective bodies and their distance from each other. There is no causal connection postulated, at least not in the sense typically used by sociologists. There is simply a statement of equivalence: one force is related to other properties of the universe. To this day, the issue of what sociological laws should look like haunts theory. We often speak in the language of causality, but at the more abstract level where empirical context is removed from a principle, causality gives way to basic relations among forces in the universe. It is not evident that Comte had a very sophisticated view of causality and its pitfalls in mind, but his words pull us into a debate that has never been resolved, either at a philosophical level or at the level of the actual practice of science.

Third, the reference to final causes is vague. He states that he means ‘purpose,’ but does he mean some ultimate goal or a function? It simply is not clear; and if he means function, he did not follow his own advice, since Comte reintroduced functional analysis into social theory.

Fourth is a clearly stated view that sociological theory will have relatively few laws, because abstract statements should be reduced to ‘the smallest number possible.’ Comte thus had an image of sociological theory as resembling the astrophysics of his time; indeed, as was also the case with Spencer, positivism emerges as an effort to emulate astrophysics in generating very abstract laws that state the basic nature of relations among generic forces in the social universe.

Yet, Comte also recognized that these abstract laws need to be applied to specific empirical contexts; and he proposed a kind of division of labor in the natural sciences and, hence, in sociology:

we must distinguish between the two classes of Natural science—the abstract or general, which have for their object the discovery of laws which regulate phenomena in all conceivable cases, and the concrete, particular, or descriptive, which are sometimes called Natural sciences in a restricted sense, whose function is to apply these laws to the actual history of existing beings. The first are fundamental, and our business is with them alone; as the second are derived, and however important, they do not rise to the rank of our subjects of contemplation. (Comte, 1854: 23)

In this passage, emphasis on the abstract, general and generic is maintained, but Comte implies something else: deductions from these abstract laws to particular cases. From its beginnings, then, positivism held to a view of theory as highly abstract but, at the same time, as amenable to translations to particular contexts.

Curiously, positivism is often equated in the contemporary commentaries with ‘raw empiricism,’ especially with the use of ‘hard’ methods and quantitative data analyses. Such a portrayal is then used to condemn positivism, but as is obvious, this portrayal is a gross distortion of what Comte intended. For he recognized that ‘if it is true that every theory must be based upon observed/acto, it is equally true that facts cannot be observed without the guidance of some theory’ (Comte, 1854: 4) Thus, empiricism is impossible since there is always some theory, implicit or explicit, guiding what one sees, but more fundamentally, empirical inquiry alone stifles the development of a true positive science:

The next great hindrance to the use of observation is the empiricism which is introduced into it by those who, in the name of impartiality, would interdict the use of any theory whatever. No other dogma could be more thoroughly irreconcilable with the spirit of positive philosophy—No real observation of any kind of phenomena is possible, except in as far as it is directed, and finally interpreted, by some theory. (Comte, 1854: v.2:242)

Thus, the portrayal of positivism as empiricism, per se, is patently false, and Comte (1854: v.2:243) went on to make sure that there was no misunderstanding about the matter:

Hence it is clear that, scientifically speaking, all isolated, empirical observation is idle, and even radically uncertain; that science can use only those observations which are connected, at least hypothetically, with some law.

In sum, then, positivism was at its very beginning a view of sociology as a theory-driven science, devoted to (1) discovering the fundamental properties that are always present when humans organize and (2) formulating abstract laws on the forces governing the operation of these properties. It was never data collection for its own sake, despite contemporary pejorative portrayals, but a view of sociology as a natural science whose goal was to develop general and abstract laws. Still, these laws need to be tested against data; and at times, the collection of data can help in the formulation of a law. Thus, Comte’s positivism was also concerned with general methodological strategies for data collection.

The Basic Methodological Strategies

While the goal of positivism is to develop the laws of human organization, Comte took seriously the methodological question of how to collect data in order to test theories and, at times, in order to induct theoretical principles from systematically collected and analysed data. Yet, his discussion of methods is rather superficial, presenting a case for four basic methodological strategies.

One strategy is observation; and here he draws upon Montesquieu’s idea of considering social phenomena as ‘things’ or, as he phrased the matter, as ‘social facts.’ When viewing the social as a thing or fact, observations stay away from biased moral judgement and, instead, focus on the statical and dynamical properties of social forces. Sociology was, therefore, to be the science of social facts.

Another strategy is experimentation. Comte did not have in mind laboratory experiments but, rather, naturally occurring situations where a pathological force interrupts the normal flow of events. Under these conditions, where the normal state of the social organism is interrupted by a pathological condition, it becomes possible to see how the more normal social processes reassert themselves in an effort to manage the pathology. Comte analogized to the physician, arguing that sociologists could do much the same thing for the ‘body social’; for just as the physician can learn about normal body functioning by observing disease, so the sociologist could understand the normal functioning of society by observing social pathologies. Of course, just what would constitute ‘normal’ and ‘pathological’ for the social organism would open up the door to moral judgement about the social world. Comte would, of course, have rejected this conclusion, but it unfortunately follows from this rather limited view of experimentation.

A third strategy is comparison, and here Comte also had a biological view of comparative anatomy (or structure and dynamics) between societies. But he also had a view of comparison between different types of social forms evident among ‘lower’ animals, as well as comparisons with past and present social forms of human organization. By such comparisons, it becomes possible to see what is similar and dissimilar and what is present and absent across various forms; and from these types of comparisons, knowledge about the fundamental properties of the social world of humans would be revealed.

The final methodological strategy proposed by Comte is historical analysis, which is a variant of the comparative method. His law of the three stages is such a historical method, examining the movement of ideas and corresponding structural arrangements across history. In looking at societies over time, Comte argued, their dynamical qualities are revealed; and it is these that will be formulated into laws of human organization.

Comte’s strategies seem rather simplistic today, but for their time, these were important insights. There is little attention to how data are to be collected within each of these four strategies, but Comte was nonetheless making a strong case for a science built upon the formulation and testing of general theoretical principles through the unbiased assessment of data.

The Substance of Sociology: Statics, Dynamics and the Organismic Analogy

Like Saint-Simon, Comte saw sociology as an extension of biology in its study of organisms. Sociology was to be the study of social organization, with an emphasis on social wholes. For ‘there can be no scientific study of society, either in its conditions or its movements, if it is separated into portions, and its divisions are studied apart’ (Comte, 1854: 225). Thus, while the basic goal of sociology was to produce laws like those in the astrophysics of his time, the subject matter was an extension of biology. Hints of this emphasis come from his interest in studying social pathologies or in comparing the anatomy of diverse social forms. In seeing the social organism as the subject matter of sociology, Comte reintroduces functional analysis into sociology: social facts are to be studied with reference to their consequences for maintaining the normal states of the social whole. Thus, positivism in sociology was originally very much married to functional analysis; subsequent positivists like Herbert Spencer and Emile Durkheim would continue this alliance of searching for laws like those in physics on a subject matter defined in biological terms.

Comte divided sociological analysis of social organisms into ‘statics’ and ‘dynamics.’ He wanted to study structure (statics), but true to his Enlightenment ideals, he also wanted to view society as progressing (dynamics). Statics is, in essence, the analysis of functions of social parts to the whole. The parts to be analysed by sociology were not individuals (these were to be the subject matter of biology), but the units organizing individuals. The ‘family’ composed minimally of husband and wife is, in Comte’s functional view, the most elementary unit of social organization, with this elementary unit becoming the basic building block for larger social units. Late in his career, long after he abandoned science, Comte [1851-54] 1875: 221-76) elaborated on this analysis, engaging in analogies between biological organisms and social organism. But in Positive Philosophy, emphasis is on examining structures in terms of (1) how various social units become collated into larger units and (2) how these larger units operate to sustain the ‘body social.’ His analysis is highly flawed, but he introduces ideas that dominate functional analysis to the present day. Social organisms are highly differentiated, and as such, it is important to know how they are held together or integrated; the key mechanisms of such integration are common morality or spirit, mutual interdependence and exchange, and centralization of power to coordinate functions. These points of emphasis were to constitute the agenda of functional sociology for over a hundred and fifty years.

Social dynamics is, unfortunately, confined to the law of the three stages in which the nature of ideas, structural forms and their modes of integration are examined for the theological, metaphysical and positivistic stages. The details of this law are not as interesting as the basic approach to social change: examine the units of the social whole; assess their modes of integration; explore the nature of the idea systems and leaders who articulate these ideas over time; examine change in the nature of units, patterns of integration, and use of symbol systems. The historical trend is for ever more differentiation of structural units and new forms of integration (i.e. power, mutual dependence, and more generalized cultural symbols). Comte even hints at the key forces driving such differentiation: increases in the size of a population and the material density of individuals (ideas that he took from Montesquieu and that were later adopted by Durkheim (1893) in The Division of Labor in Society).

Advocacy and the Reconstruction of Society

Comte’s positivism always contained a basic line of advocacy: science is superior to any other system of thought for examining the structure and dynamics of society; laws of these dynamic properties can provide the tools for reconstructing society. Sociology is the ‘queen science’ because it has been the last to go positivistic, but with its emergence, all domains of the universe can now be examined scientifically. As a result, the laws of the universe—physical, biological and social—can be used to make a better society. Such is, of course, the essence of the Enlightenment project, but it begs an important issue: who is to decide, and in terms of what moral premises, how the laws of science are to be used to construct what type of universe? Comte simply assumed, it appears, that the laws themselves would inform policy-makers of the proper direction of the social order. Obviously, this is a very naive position, but the critical point is that early positivism always contained a vision of laws being used to reconstruct the social world.

From its beginnings, then, positivism had an engineering component, or if one prefers, an emphasis on social practice. The laws of human organization were not just to be discovered for their own sake; they are to be used and applied to problematic conditions. Ironically, sociological practice as it has evolved in sociology over the past one hundred years often mounts critiques of positivism, whereas in fact, the thrust of positivism was always to use general laws for engineering applications.

Comte in Retrospect

In many ways we hold Comte in more respect today than his colleagues did at his death. Yet, we still view Comte in a kind of bemused fashion, as one who postulated the law of the three stages, the hierarchy of sciences with sociology as the queen science, and the now often rejected view that sociology could be a true natural science. We can ask if this is any way to treat a founder, but there can be little doubt that the founder of sociology is not highly regarded today. There is, of course, some basis for this low regard, but his advocacy for positivism remains sound. Sociology can be a natural science; the subject matter of this science is social structures (statics) and social processes (dynamics); the goal of sociology is to develop abstract general laws on the forces that explain the operative dynamics of this subject matter; these laws should be constantly assessed against the empirical facts; and the verified laws of sociology should be used in engineering applications. Comte’s positivism is simply an advocacy for what all scientific disciplines do; our retrospective view of Comte as a flawed figure comes not only from the fact that Comte was indeed an odd man, but also from contemporary sociology’s ambivalence over its scientific prospects.

The Origins of Herbert Spencer’s Positivism

In 1864, almost a decade before Herbert Spencer entered sociology, he published an article titled ‘Reasons for Dissenting from the Philosophy of M. Comte.’ In this essay, Spencer (1864) stressed that he disagreed with Comte on the following issues: that societies pass through three stages of development; that causality is less important than relations of affinity in stating laws; that government can use the laws of sociology to reconstruct society; and that psychology is merely a subdiscipline of biology. He did, however, agree that knowledge comes from observations of facts and that laws about the invariant properties of the universe could be formulated. Spencer also accepted Comte’s label for the discipline, sociology, and he gave Comte credit for reintroducing the organismic analogy back into social thought, although he was quick to point out that Plato and Hobbes had made similar analogies and that von Baer had greatly influenced Comte’s views. This article almost reads as if Spencer is ‘protesting too much,’ but it is hard to know for sure since at the time Spencer had not yet written any of his major sociological works. But like Comte, his sociology was to emphasize the search of laws of the universe and to employ a mode of functional analysis inspired by an organismic analogy. And so, to the extent that Spencer can be seen as the carrier of positivism from the 1870s to the turn of the century, sociological laws in Spencerian sociology were (1) to be deduced from the laws of nineteenth century physics and (2) to address a subject matter conceptualized as ‘super-organic’ systems.

Spencer saw himself as a philosopher, and his goal was to develop what he termed a Synthetic Philosophy that encompassed all domains of the universe, social and natural, and including morals. There was, then, always a moral component in Spencer’s work, and indeed, our retrospective appreciation of Spencer is diminished by his ideological biases that are considered highly conservative today but, in fact, were very liberal in Spencer’s time. If we ignore the moral works, which mark the beginning and very end of Spencer’s career, we can see that he converted Comte’s crude organismic analogy and simple view of scientific laws into a highly sophisticated sociology. Like Comte, this sociology was very much influenced by the Newtonian revolution, but equally significantly, Spencer was influenced by the emerging biological sciences indeed, he wrote one of the major treatises on the topic. He had read Thomas Malthus (1766-1834), who was not, of course, a biologist but a political philosopher whose influence on biology was none the less profound; and more directly in biology, he read William Harvey, Ernst von Baer and Charles Darwin. Moreover, Spencer associated in London clubs with the most eminent scientists of his time, a good portion of whom were biologists. Thus, by the time Comte’s career had come to an end, Spencer had spent decades absorbing real science; and this greater familiarity with the advances in science during the decades of the nineteenth century enabled Spencer to develop positivism on both substantive and methodological grounds.

The rather strange vocabulary of Spencer’s model of human evolution is partly the result of his effort to subsume all the sciences, and ethics as well, under what he termed ‘first principles.’ These were published in 1862 in a book by this name, and the goal was to delineate the cardinal or first principles that govern the operation of the universe. These are borrowed from the physics of his time and concerned such issues as (a) the indestructibility of matter, (b) the continuity of motion in a given direction, (c) the persistence of force behind movement of matter, (d) the transferability of force from one type of matter to another, (e) the tendency of motion to pass along the line of least resistance, and (f) the rhythmic nature of motion. All other principles relevant to a particular domain of the universe—physical, chemical, biological, psychological, sociological, or ethical—could be deduced from these first principles. The details of translating these ideas to specific areas of inquiry are less interesting today than the intent: to create a unified general systems theory for all domains of the universe. This unity came from highly abstract—and obviously rather vague-ideas about matter, motion and force. Spencer’s positivism was thus even more grandiose than Comte’s because he believed that the same laws could be used to understand every realm of the universe; all that was necessary was to translate these first principles and, then, add necessary refinements as inquiry into a particular realm of the universe is undertaken. By the time Spencer was actively engaged in sociology during the 1870s, he had written not only First Principles (1862) but also The Principles of Psychology (1855-1872) and The Principles of Biology (1864-1867).

The Positivism of Herbert Spencer

Spencer’s Methodological Work

Spencer’s sociology can be broken down into two major components, one methodological and the other substantive. Spencer began his foray into sociology with two important methodological efforts. One was the publication of The Study of Sociology in 1873, and the other was the compilation and collation of comparative historical and ethnographic works into Descriptive Sociology. By the time that The Study of Sociology was published, work on Descriptive Sociology was well under way, with the first volume published in the same year as The Study of Sociology. The multiple volumes of Descriptive Sociology that were to appear over the next sixty years represented an effort to pursue the comparative method advocated by Comte by assembling data on what was then known about populations all over the world, both literate and pre-literate. For literate populations with a written history, Spencer hired professional scholars to compile a history of a society; for pre-literature, he employed scholars to compile the data from anthropologists and travelogs on diverse populations. The goal, as Spencer noted in the preface of Volume I on The English (1873: vi), was to provide data for The Principles of Sociology:

In preparation for The Principles of Sociology, requiring as bases of induction large accumulations of data, fitly arranged for comparison, I commenced by proxy the collection and organization of facts presented by societies of different types, past and present… the facts collected and arranged for easy reference and convenient study of their relations, being so presented, apart from hypotheses, as to aid all students of social science in testing such conclusions as they have drawn and in drawing others.

Thus, Spencer’s positivism was to rest, like Comte’s advocacy, on ‘social facts’ induced from the data available on diverse populations; and the analysis of super-organic bodies was to be comparative, examining different types of societies and, it should be added, different species of animals that are organized in terms of a division of labor (that is, the social insects). Unlike Comte, who merely advocated a comparative methodology, Spencer executed it. In all of the volumes of Descriptive Sociology, a common category system is used to list facts. These categories allow for comparison of one society across its history or for the comparison of different societies in the past or present. And as Spencer emphasizes in the quote above, these comparative data formed the basis for induction of theoretical principles. Thus, for Spencer, as one moves from highly general ‘first principles’ to specific realms of the universe, it becomes necessary to array the data in systematic ways. With the array of data, generalizations can be inducted that can help develop the more abstract laws connecting the ‘first principles’ to the social world. In Spencer’s sociology, then, deduction and induction are critical; they both can facilitate the formulation of general theoretical laws that, on the one side, apply the first principles and, on the other, make them sufficiently concrete so as to explain the operative dynamics of a particular subject matter.

The Principles of Sociology is such a long work because it is filled with data. The actual theoretical statements take only a few hundred pages; the rest is example after example from Descriptive Sociology as well as findings from the other sciences, particularly the data assembled for The Principles of Biology. Thus, Spencerian positivism is highly theoretical, but it seeks to formulate theories that have been disciplined and assessed by social facts from a wide variety of sources. For if laws are to be truly general and universal, Spencer appears to have argued, they must explain the data from a wide range of specific empirical cases.

The other major methodological work in Spencer’s positivism is The Study of Sociology, which represented both a call to scientific sociology and an effort to quiet critics of social science in general. In this work, Spencer was to address not only the sources of bias inherent in humans studying human society, but perhaps more fundamentally, he anticipated many of the criticisms against efforts to develop general scientific laws in sociology. For those who would argue against the existence of fundamental forces directing human organization, Spencer argued that policy-makers and lay persons alike constantly make this assumption when they presume that their remedies for social ills will indeed solve problems; otherwise, if they did not feel that the social world had basic forces that could be shaped to their will, they would not be so adamant in their advocacy for particular policies and programs.

What lay persons implicitly assume, sociologists must explicitly pursue. Human organization is guided by generic forces that operate in a lawful manner, and ‘it behooves us to use all diligence in ascertaining what the forces are, what are their laws, and what are the ways in which they cooperate’ (Spencer, 1873: 47). For those who argue that sociology cannot be an ‘exact science,’ like the natural sciences, Spencer countered that many of the insights of the ‘hard’ sciences are stated verbally and that the methods used by researchers in these sciences are often qualitative. For when scientists must work in natural systems and are, therefore, unable to measure precisely variables nor to control for their interaction effects, it becomes impossible to engage in purely quantitative analysis. But, these constraints do not make research or theory any less scientific, nor do they make laws less powerful. For it is evident that in many situations, ‘factors so numerous and so hard to measure, that to develop our knowledge of their relations into quantitative form will be extremely difficult, if not impossible’ (Spencer, 1873: 45).

In these opening passages of The Study of Sociology, Spencer thus removes the burden of quantification from positivism. The goal is to isolate the forces of the social universe, state their operation in laws, and seek to understand their relations to each other. Such activity need not be stated as a mathematical equation, as in Newton’s law of gravity, nor do the data collected to assess the plausibility of a law need to be quantitatively measured. Additionally, sociologists should not use prediction of events as the criterion of a science, since in complex natural systems, such efforts become difficult; instead, the criterion of all science should be: Do the laws on the basic forces of the universe lead to an understanding of why an event occurred? When an affirmative answer can be made to this question, it then becomes possible to have a scientific explanation for a specific empirical event, even if these events could not be predicted with any precision.

The rest of The Study of Sociology addresses sources of bias that need not be reviewed in detail here, except when Spencer’s detailed discussion bears on the nature of positivism. Spencer argued that the data collected in science should be directly relevant to formulating or testing the laws on the timeless forces that govern the social universe. Research problems selected for other purposes—for example, the desires of benefactors and funding agents, the dictates of public opinion, the commitment to a research technique or research paradigm, the dominance of particular agents and programs, and the social positions of researchers themselves—should not distort the collection of data away from the fundamental, basic, generic and universal forces governing the operation of the social world. Futhermore, the collection and analysis of data should not be biased by a cherished hypothesis and ideological commitments which keep researchers from assessing objectively and critically the plausibility of their theories. Finally, it is important to collect data over time in order to see processes unfold rather than to take only a cross-section of data that does not give a sense for process. If all these sources of bias can be overcome, then it is possible to have a true science of society.

Spencer’s Substantive Analysis of Super-Organic Bodies

Spencer is most famous for his organismic analogy and functionalism, and as a result, early positivism is somewhat tainted by this association with what some see today as a discredited approach. But, if we take literally, as Spencer did, the title of his work ‘The Principles of Sociology,’ functional analysis on super-organic bodies was designed to produce abstract laws of human organization. Added to this functional analysis was an evolutionary approach in which Spencer traced the long-term development of human societies from hunting and gathering to industrial forms of organization. This too, for a time, stigmatized positivism as it became associated with models of evolution which, by the early twentieth century, were in fast decline. Yet, if we look more closely, Spencer’s heuristics for analysing society—that is, functionalism and evolutionism—were designed to generate abstract laws of human organization that followed from his ‘first principles.’ Detailed analyses of the sociological principles derived from the first or cardinal principles and induced from the data arrayed in Descriptive Sociology can be found in a number of places (for example, Turner, 1984; 1985; 1998: 74-7; 2000), and so they need not be enumerated here. Instead, we can simply summarize the basic intent and substantive thrust of the principles.

For Spencer, long-term evolution involved increasing differentiation of a population and the structures organizing the activities of this population. The basic cause of this differentiation was population growth which placed escalated logistical loads on basic social functions: production, reproduction, distribution, and regulation. That is, as populations grow, they generate selection pressures for new ways to expand production, to assure adequate reproduction of human capital, to distribute goods, resources, services and information, and to coordinate and regulate the increased number and expanded volume of societal activity. Those populations that cannot meet these new challenges will face dissolution and will, therefore, de-evolve, whereas those that do, will become more differentiated with respect to the economic division of labor, the diversification of reproductive structures, the extensiveness of transportation and communication infrastructures as well as market exchanges, and the consolidation of political power. These are, in Spencer’s eye, fundamental forces of human organization; and their interrelated dynamics explain much of what occurs in super-organic systems. These forces also set off certain dialectical dynamics. One of these is growing inequality associated with expanded production and concentrated power; and as inequality increases, internal threats increase and place new logistical loads on regulatory structures. As threats mount, ever more power is concentrated which only ratchets up the level of inequality and escalates the threats to new levels, eventually placing enormous disintegrative pressure on the society. Another dialectical dynamic is the external threat from the geopolitical arena, which increases the concentration of power which in turn only serves to escalate the level of inequality and, hence, the potential for internal threats. Moreover, as more territory is conquered, the level of inequality increases as does the diversity of the conquered populations, thereby increasing internal threats. All these ideas are developed as an abstract series of first principles, and they constitute some of sociology’s most important laws of human organization, although most contemporary sociologists often do not recognize that Spencer was the first to articulate these basic laws. Spencer also engaged in a more detailed analysis of institutional systems, applying these general laws and, at the same time, producing lower-level generalizations about institutional dynamics during the course of human evolution.

Thus, in contrast to Comte who had only advocated a search for general laws and their application to the dynamics of social organisms, Spencer actually executed the strategy, generating some of sociology’s basic laws about the relations among the fundamental forces of human organization: population, power, production, reproduction, distribution, geopolitics, inequality and conflict. These laws were illustrated with many examples drawn from Descriptive Sociology, and so, Spencer’s positivism was both a highly abstract theoretical exercise at discovering fundamental laws and a comprehensive effort to illustrate the plausibility of these laws with empirical examples drawn from a wide range of sources.

The Transfiguration of Positivism in the Twentieth Century

To the extent that we consider Comte and Spencer the founders of sociological positivism, it is clear what they advocated: search for the general laws of human organization and assess these against empirical facts. Positivism, in their eyes, is simply the equivalent of producing general theory, through derivations from other theoretical principles and through inductions from empirical generalizations, or both. This was positivism, plain and simple.

One of the more interesting historical questions is how positivism became associated with either raw empiricism and data analysis employing quantitative methods or, in the eyes of critical theorists, with antihumanism legitimating the systems of power in a society and the status quo. Comte and Spencer clearly did not advocate the view of positivism as mere data collection, and Comte obviously saw positivism as supporting humanistic ends and goals, and if necessary, as challenging the status quo. Even Spencer was highly critical of concentrated power, and he argued strongly against the use of power to extend a society’s geopolitical boundaries. In closing this review of early positivism, then, it might be useful to examine this transfiguration of positivism to connote in the minds of many an approach so different from what Comte and Spencer championed.

The Vienna Circle is the key to the change in positivism; and although Ernst Mach (1893) was not part of this circle, he is still a key figure in the circle’s deliberations. Mach had argued that science should not speculate on unobservables, rejecting the kinds of ‘natural laws’ postulated by positivists. Instead, theory should be mathematical descriptions of immediate sense data. This line of argument framed much of the debate within the Vienna Circle which, for the most part, was concerned with logic and systems of formal thought. But soon, in light of Mach’s advocacy, a split developed over the relative merits of data or logic in generating understanding of the universe. The radical faction argued that truth can be ‘measured by logical coherence of statements,’ whereas the more moderate group stressed the need for a ‘material truth of observation,’ supplementing any ‘formal truths’ generated by logic (Johnson, 1983: 189). Karl Popper (1959, 1969), who was a somewhat marginal figure in the Vienna Circle in the 1930s, was the most famous mediator of this split, suggesting that a formal theory can never be proven and, therefore, must be constantly subject to empirical tests to sustain its plausibility.

How did this debate within the Vienna Circle transfigure the meaning of positivism? The answer resides in America, where there was a compulsive concern with sociology’s status as a science in the early decades of the twentieth century (Turner and Turner, 1990). Both Mach’s and Popper’s arguments were appealing, because they legitimated the kind of variable analysis (for example, cross tabulations and the beginnings of Pearsonian correlations) that was emerging in American sociology in the late 1920s and 1930s. By the 1930s, sampling, scaling and statistical analysis were increasingly becoming mainstays of the discipline. The Vienna Circle had, in the meantime, invented a redundant term—logical positivism—to describe the process of deducing hypotheses from abstract theoretical formulations which, for Comte and Spencer, was simply positivism. But this division between logical deduction on one side, and empirical generalization on the other, gave legitimation to quantitative data analysis that was to test hypotheses deduced from abstract laws; and since no theory is ever proven (à la Popper), it must constantly be assessed with quantitatively analysed data. Thus, the collection of data and their quantitative analysis were essential to carrying out the mandate of ‘logical positivism.’ This position was not very different from that argued by Comte and Spencer, but the emphasis on quantification went against Spencer’s advocacy and, more significantly, began to convert positivism toward the collection and analysis of data per se, without too much regard for theory or for the ‘logic’ of deduction.

Ironically, in the 1970s, a theory construction movement swept over American sociology with the production of texts on how to ‘build’ and ‘construct’ theory. Here was logical positivism converted into ‘cook books’ about procedures for building theory, but these books were too mechanical. They sought to do for the ‘logical’ part of positivism what statistics textbooks had done for data collection and analysis; and they soon fell into deserved obscurity. With their demise, the last remnants of the theoretical part of positivism gave way to quantitative sociology in the United States and as a result, positivism increasingly became associated with sampling, use of scaling techniques, statistical analyses of various sorts. The emergence of the computer, especially the desk-top computer, only accelerated a trend that was well under way in the late 1950s. Today, we are left with this legacy in the United States, although in Europe positivism is better appreciated for what its founders—Comte and Spencer—advocated. Still, in Europe, there tends to be considerable skepticism about the prospects for a true science of society, and so in a somewhat different way, positivism is often perceived in negative terms.

Thus, the promise of positivism trumpeted by Comte and Spencer has given way to considerable skepticism. Faith in the Enlightenment project is constantly assaulted by various waves of anti-science rhetoric, most recently by postmodernism. In this postmodern context, positivism is viewed as a failed epistemology, or as a misguided effort to create a Grand Narrative. And, of course, in the more American context, the association of positivism with compulsive quantitative data analysis offers another line of criticism. And for critical theorists in Europe and the United States, positivism is seen not only as inhumane but as an instrument of domination in which science is used to legitimate the status quo and current systems of power. Comte and Spencer would probably turn over in their graves if they could see how their vision has been so transfigured.

Still, despite these alterations to the original meaning of positivism, there remains in the discipline of sociology as a whole a belief that theory should guide research; and while this idea is rarely executed in practice, there is a sense that data should have more general theoretical implications. This connection between data and more general theoretical laws is the essence of positivism, and even if it is only an ideal that is only occasionally implemented, most sociologists today would at least confirm their faith in this ideal. Thus, even with the transfiguration of Comte’s and Spencer’s meaning, the essence of positivism lives on. Perhaps it is no longer called positivism, given the unsavory connotations of this term in the modern eye, but it is the spirit of what Comte and Spencer saw as the essence of sociology.