If you asked a biologist, “are there stages of hominid evolution?”, they’d look at you like you were crazy. Of course, biological evolution is, from a teleological standpoint, directionless. So, why has sociology been so preoccupied with classificatory stage models? Indeed, the biggest black eye the social science has when it comes to evolution is the stage models…And, worse, they remain pedagogical tools for easy classical theory lessons, which means they are unconsciously reproduced as both factual theoretical frameworks and reasons for rejecting evolutionary principles in the social sciences.

First, some context. In the nineteenth century, Darwin’s theory was in its infancy, but its implications were immediately understood as explosive. It is not shocking that sociologists and anthropologists quickly adopted some of the language. The modern synthesis between Darwinism and Mendel’s genetics, however, was half a century or more away, and thus, it is also not shocking that these earliest efforts were failures in many regards. What masqueraded as scientific analysis of social evolution was, often, in fact veiled racism, colonialism, ethnocentrism, and white supremacy. Not always, but in, say, one of the more simple and common stage models (savages > barbarians > civilized), it was difficult to interpret any other way (Morgan 1877; Veblen 1899). But, how and why these models returned in the 1960s (Parsons 1964; Bellah 1970) is much less understandable.

In what follows, I review some of the more common modeling techniques or models and use them as examples for what is wrong with modeling, why we need to stop teaching it even if it keeps our lessons true to the theorist’s own work, and, finally, what sociology can do to build multi-linearity into its evolutionary theories.

The Two-Stagers

Every sociologist has learned, at some point, that there are two types of societies: Gemeinschaft and Gesselschaft; mechanical and organic; primary and secondary. That there are communal and associative types of relationships and collectives, and that pre-industrial societies tended to be dominated by the former and not the latter is a fact. But, what is less clear is whether the premodern/modern, or traditional/modern dichotomy so ingrained in contemporary US sociological ideology is real. To be sure, there are major differences, in size, scale, and complexity, between foraging societies and the US or Canada; there are differences worth highlighting between agrarian polities 5000 years ago and the western-style democracy; and, there are differences between religious, legal, educational elements of medieval Prussia, Sung China, and medieval Islam and today. But, are they dramatic enough to posit a two-stage model in which we (so-called moderns) are different from premoderns?

In a previous post, I remarked on the fact that human anatomy and cognition has not evolved dramatically over the last 50,000 years meaning we are, in essence, the same animal today as before. I do not want to overstate this fact, as it is obvious that fossil fuel technologies have changed the human condition. It is also obvious that humans this hardware humans carry was not evolved for large, depersonal urban societies (Maryanski and Turner 1992). However, and perhaps ironically, the types of social relationships modern urban humans have (a few strong ties and many weak ones) is actually more in line with the earliest human societies (Turner and Maryanski 2015; Maryanski 2018). A point that actual grinds against conventional wisdom that humans are “bowling alone” and society has totally broken down (Putnam 2001; McPherson et al. 2008).

If we step back and see these two-stages as Charles Cooley did, then they are useful. They denote, formally, two types of social relationships. But, the harm has been done if you were to ask me. Critical accounts of modernity, ranging from the more critical ones like Marcuse’s one-dimensional man to those less intense like Bourdieu’s simply assume there is a premodern and a modern time. And while the former explicitly romanticizes the “old days,” most sociologists rarely bother to read ethnographies or history such that they actually know what the old days were like.

The Unfolding “Have-Tos”

The worst offenders are the models that assume, explicitly or not, that society has evolved in some linear set of stages. Karl Marx’s model is the classic example (Primitive communism > slave > feudalism > capitalism > socialism > communism). There are many fundamental problems with this model beyond its teleological bent and its strange materialist Hegelian bent (humans fell from Eden and are working to return to a heaven on Earth). First, the data available to Marx was not great, but his idea about primitive communism is a bit of an embarrassment. Second, the first stage evidently encompasses 250000 BP to 5000 years ago. It covers a tremendous amount of variation beginning 10000-12000 BP when humans began settling down en masse. The slave period then lasts 2500 years until the medieval period in Europe. To compensate for its blatant eurocentric perspective, Marx adds an Asian pathway; whatever that is. It does far more to obscure evolution than provide clarity and rigor.

Perhaps its worst offense is the idea that evolution is linear. This idea rests in most classical evolutionism, predicated on some sense of progressivism (societies are getting better – e.g., more civil, more moral, smarter, etc.) and/or the idea that evolution always proceeds from simple to complex in a straight line. The first position is colonialist at worst, and misplaced idealism at best. This is not to say Weberian pessimism is the better mode, but Marx’s belief in a final Utopian stage lacks complete understanding of past human societies, human nature, or even other animal societies. More generally, however, convergent evolution (or independent cases that converge on structural and cultural similarities) is more rare than divergent evolution (cases with the same conditions going in different directions). This historical tendency was echoed in Israeli sociologist, Shmuel Eisenstadt’s (1964, 2000) work on multiple modernities. 

The second position is also poorly conceived, as it fails to recognize that some epoch-making moments have been about simplification. Protestantism, and almost every form of mysticism has been about dedifferentiation in organizational and symbolic content. The former, regardless of one’s stance on the famous Weberian Protestant ethic thesis, represented a huge change in western society.

The Exceptional Cases

Marx’s Asian pathway alludes to a third weakness: the exceptional case models. Robert Bellah’s (1970) religious evolution model represents this perfectly (Primitive > Archaic > Historic > Early Modern > Modern). Besides the poor choice in labels, the Early Modern stage represents a single case: European Protestantism in the 17th century. The logic here is confusing: how can a single case become a whole stage? Particularly when no other religion has met that criteria and most skip to modern from either archaic or historic?

Bellah’s model is, admittedly, partially a Weberian historical ideal type, and thus not, like Marx, an serious belief in compulsory stages. But, the model itself raises so many more questions that it usefully answers. For instance, Judaism, Buddhism, and Hinduism are “Historic” religions both in terms of when they evolved and in terms of the purest literal sense of the word. But, how can a religion be fixed in a stage when it currently exists in modernity?

These questions underscore the ambiguity stage models present for units of evolution: is it religion(s)? Society? Religious symbols? Organization? In part, sociologists in both the classical and many in the more recent 50 years have failed to specify what it is that is evolving; at least until recently.

Lenski and Service

Not all stage models are bad in their entirety. When social scientists classify societies for comparative analysis and in Weberian fashion, there is some utility. Elman Service’s (1971) famous political sovereignty model (Band, Tribe, Chiefdom, State) and Gerhard Lenski’s (1970) subsistence technology model (hunter-gatherer; horticultural; pastoral; agrarian; intensive agricultural; industrial; post-industrial) are instructive. The former groups societies into classes that share similar political mechanisms of organization, while the latter emphasizes the primary mode of subsistence. Both have considerable impact on other institutional spheres (e.g., agrarian societies tend to have priesthoods and palace-economies whereas horticulturalist do not have the surplus for status markets as well as enough food to support urban life).

The strength, here, is built into the comparative lens. Like a biologist seeking to create taxonomies that are intentionally un-hierarchical, these models provide a common language from which scholars can refer to clusters of similar, but not identical societies. Moreover, they do not presume a single causal explanation, nor linearity. Chiefdoms are, as both the archaeological and ethnographic record indicate, often evolutionary “dead ends” (Abrutyn and Lawrence 2010). And, more than a fair share of tribe-level societies have “skipped” the chiefdom stage and leapt right into state-like societies. Likewise, Lenski’s model integrates diffusion as one mechanism typically missing from stage-models. Clearly, a pastoral or simple agrarian societies can bring “foreign” technology into its economic system, either by colonization or through trade, adaptation, and so forth. There is no reason to believe a society can transform, over one generation, from one type of subsistence economy to another. The former Soviet Union attempted to jump from agrarian to industrial, with varying degrees of success.

What’s Missing?

There are several angles one could take in positing what is missing from stage modeling. But, I would rather focus on what is both most and least obvious: the general lack of anything beyond the most macro-level sociology. To be sure, I am an ardent proponent of general models that emphasize highly generic processes, which is best specified at the very macro in terms of temporality and complexity (Lenski 1988; Turner 1995). Sociologists would be fooling themselves if they did not think population pressures, resource scarcity, heterogeneity (broadly defined), and natural/social exigencies were not powerful (if not invisible) forces across time and space. The current climate change issue is primarily about these.

But, these generic processes often obscure the ways sociology, dropped down a few levels, can contribute to understanding how and why sociocultural evolution happens the way it does. In a little cited text (in sociology at least), Richerson and Boyd (1992) make a case for focusing on the “micro” or specific evolutionary processes by which much of history is made. By micro, they are referring to smaller snippets of time and more meso-level events, like epochal social movements. Here, as I have done previously, they make a case for Lamarckian evolution at the cultural level – that is, purposive, sometimes quite rapid evolution.

How I interpret their argument is as such: sociology has a massive set of theoretical and methodological tools for studying how collective behavior can reshape societies towards reform and, in rarer but no less important cases, revolution. At the meso-level, we can discern in far greater details the mechanisms, contextual factors, and other relevant bits and pieces for thinking more generally about how and why structure and cultural evolves. Too often, the stage-models which are simply a symptom of a much larger malaise of thinking too macro and general all of the time, stand in for good explanation. In doing so, they commit the same errors – albeit much less egregiously and with far more nuance and empiricism – that Marx did in reproducing Hegel’s own dialectic: history appears to unfold, inevitably, with little human intervention. And, if there is human intervention, it is driven by history itself. But, historical-comparative sociology teaches us, if anything, that whilst the great man theory is bunk, the great group theory is far more useful in understanding both successful evolution (that is, actual change) and failures (e.g., bad decisions that led to societal collapse or disintegration).

Thus, in my final post on this subject, for now, I will hone in on general v. specific evolution, to highlight the importance of both for evolutionary sociology.