ADAM SMITH AND ECONOMIC DEVELOPMENT IN THEORY AND PRACTICE: A REJECTION OF THE STADIAL MODEL?

Abstract

Adam Smith allegedly offers a model of economic development in both his Lectures on Jurisprudence and the Wealth of Nations—the so-called four stages of development model. The model presents a linear unfolding view of economic development from primitive to advanced stages. But Smith’s own historical examples systematically contradict this model. I thus question whether Adam Smith actually endorses and uses the four stages model of development to illustrate development and suggest that if he does, he does it to discredit it instead. For Smith, history teaches that development is more accidental than fitting deterministic models.

I.

Considering the stadial model of development as a model of economic development in Adam Smith independently from the historical empirical evidence he offers may be problematic. Smith allegedly offers a clear stadial model of development both in his Lectures on Jurisprudence ([1762–63] 1982) and in the Wealth of Nations ([1776] 1981). Yet, Smith also presents lists of historical examples that contradict his alleged model, perhaps questioning the validity of the model instead of endorsing it, if he does indeed use that model at all.

I first look at the alleged model of the four stages of development as presented in the Lectures on Jurisprudence, then at some of the historical examples of development in the Wealth of Nations, namely, the development of towns as well as the development of early civilizations and of late colonial development. All the examples Smith uses are exceptions to his allegedly theoretical model, a strange choice if Smith really wanted to use that model to explain development.

What I will not do is to formalize Smith’s model, as Walter Eltis (1975) and Samuel Hollander (1973) have done for growth, but stay with Smith’s narrative. I will not take part in the debate on Smith’s materialism (Pascal 1938; Meek 1976; Skinner 1982) or lack of it (Winch 1983a; Haakonssen 1981), about which John Salter (1992) provides a review, but I will look at Smith’s use of stadial models of development and the examples he uses. While I agree with Hla Myint (1977) that Smith has a “trade–cum–development” theory, I differ from Myint (1958) when he looks at the validity of Smith’s model with actual empirical facts. I focus exclusively on what Smith himself offers as empirical data, and not on whether his historical account of development is correct. Additionally, I agree with Paul Bowles (1986) that the disconnection between theory and facts is a problem, but I do not share his view that this is a failure on Smith’s part due to the influence the Physiocrats had on him. If Smith failed to establish “that the ‘natural progress of opulence’ described the way in which economies would, if left to themselves, develop, and that his failure … results from his methodological approach and, in particular, from the fact that the ‘natural progress of opulence’ was an a priori concept formulated without reference to historical evidence” (Bowles 1986, p. 110), why would Smith be bothered to continuously list historical evidence that contradicts his model? I thus raise the question as to whether Smith himself was questioning the use of the stadial model as a model of development.

The particular form of the model of stages of development I consider implies that “societies undergo development through successive stages based on different modes of subsistence” (Meek 1976, p. 6; emphasis in original). The key to this stadial model is thus not just that the modes of subsistence determine the stages of societies but that there is development from one stage to another, that there is progress in society, even if it is possible that different modes of subsistence coexist at the same time. This is the crucial aspect of Ronald Meek’s interpretation—that “development should be regarded as proceeding through four normally consecutive stages” (Meek 1971, p. 10)—which, in my reading, Smith either does not adopt, or, if interpreted as adopting, I suggest Smith is criticizing rather than endorsing. In this I agree with James Alvey (2003) that teleological interpretations of Smith are problematic. I differ from Alvey as I do not think “Smith was confused” (Alvey 2003, p. 19), as I suggest that Smith either does not adopt a model of stadial development as a model of development, or, if he uses it, he does it to criticize it, not to endorse it.

Indeed, I take issue with Meek and Meek-like interpretations that see, as Meek would say, that “in the beginning all the world was America … America … is a pattern of the first ages in Asia and Europe” (John Locke, cited in Meek 1976, p. 22)—the idea of stages of development that implies a more or less “orderly sequence or succession of different modes of subsistence through which societies could be conceived as progressing over time” (Meek 1976, p. 23) even if different societies may at the same time be at different stages. This does not mean that I do not see development in Smith. Development is an historical fact in Smith. But the way in which Smith explains this historical fact is not through a stadial model of “orderly sequence or succession of different modes of subsistence.” This does not mean, either, that Smith does not consider different stages of society. He does, of course. But, following Christopher Berry (1997, 2013), I suggest that the stages of society in Smith are simply a pedagogical heuristic, a classification of different kinds of society, and in a sense not dissimilar from Aristotle’s (1877) classification of hunting, pastoral, and agricultural societies in Book I of Politics. The four stages are a taxonomy of different relations between means of production and social, moral, political, and legal institutions, not a model of development from one stage to another. It may not be an accident that Smith tends to use “state of society” rather than “stage of society,” where “stage” is a point, period, or step in a process or development, while “state” refers to the particular condition that someone or something is in at a specific time. ¹

That Meek-like interpretations face problems is not new. Contra Meek, Smith fails to present any data in support of the actual stadial development of societies. To the contrary, his supporting data contradict stadial development. And yet, there seems to be a reluctance to abandon this idea of inevitable progress in Smith (cf. Campbell and Skinner 1976; Winch 1983b; Haakonssen 1981; Evensky 2005). Even if one takes the softer interpretation of Harro M. Hopfl (1978) that “conjectural history is concerned with the typical society … and its ‘natural’ (that is to say, typical) progress through the stages of advancement, there need be no correspondence between the natural course of progress and the actual ‘empirical’ history of a particular society, for the latter might be fraught with ‘accidents’” (Hopfl 1978, p. 31), we run into the problem of Smith offering no empirical support, and the one he offers goes in the opposite direction. Similarly, James Alvey (2003, p. 20) concludes, “It seems to be an impossible task to make Smith consistent other than by ignoring some passages in his work.” But then he tries to save a teleological reading of Smith, saying, “Perhaps we can conclude by proposing that Smith is an 80-per-cent optimist” (2003, p. 20; “optimist” because of the telos of human progress). On the other hand, the problem of inconsistency does not arise if we think that having different kinds of societies does not imply that one will develop into another, or that the inconsistency is presented to question the correctness of the model.

If one really wants more than just a classification system from the stages of society in Smith, if one really wants to attribute the presence of a stadial model of development to Smith, which may be questionable, a possible alternative way of reading Smith may instead be that Smith knew well that the four stages could be used as a development model à la Meek, but he could not get himself to support it. Rather, he offered an alternative. After all, he tells us, we are naturally driven to build models, to build systems, to explain an otherwise puzzling reality. So, we observe nature and to alleviate our anxiety at seeing all these disconnected events, we imagine “invisible chains which bind together all these disjointed objects … [thus] … introduc[ing] order into this chaos of jarring and discordant appearances” (Smith 1982, “Astronomy 12,” pp. 45–46): we develop theoretical systems. But these models are the products of our imagination; and they are not necessarily correct or definite. So, in his essay “History of Astronomy,” Smith goes through the history of astronomical systems—models of the skies meant to link together different phenomena. But when a system is unable to deal with empirical evidence that does not fit, eventually that system is replaced with a different one (on different ways of evolution of knowledge within economics, see Khalil 1995).

In the Wealth of Nations, we have the description (and the criticism) of two economic systems: the mercantile system and the agricultural system. Both systems offer a more or less coherent way to connect the dots of economic phenomena with invisible chains. Yet, these systems, like the astronomical systems, are meant to be replaced by other systems better able to explain more otherwise unexplainable things. Allan D. Megill (1975), I think correctly, notes that Wealth of Nations is not a system of political economy but an inquiry into the nature and causes of the wealth of nations: an inquiry, not a system. In this light, one may read Smith’s account of economic development not as an endorsement of a Meek-like stadial theory but as a criticism of it. Whenever there are things that violate the order we expect to find, we are surprised and we wonder. We try to reconcile them and to smooth them into that invisible chain that we imagine being present. But eventually too much evidence contrary to the current system will lead us to look for a different way to connect those dots. When the empirical data do not fit our model, we eventually start looking for a different one, like we did with astronomy. So, when none of the empirical data fits our stadial model of economic development, maybe it is time to inquire again into what causes nations to develop and grow richer.

II. LECTURES ON JURISPRUDENCE

The clearest exposition of Smith’s stages of development comes from the set of his students’ notes complied into the Lectures of Jurisprudence ([1762–63] 1982). The Lectures on Jurisprudence (LJA) were found well after Smith’s death and thus not published by Smith or during Smith’s time. Given that Smith asked his executors to burn all his non-published writing before he died, and that the lectures are students’ notes, not Smith’s own notes, some may hesitate to use them with the same weight as Smith’s approved published works. The Lectures on Jurisprudence are generally considered useful, though, because they contain, among others, earlier drafts of what would become the Wealth of Nations ([1776] 1981).

In these lectures we can find what is known as the “four stages of development.” Stage theory was a common construct of the Scottish Enlightenment, as Craig Smith (2006) demonstrated (cf. Berry 2013; see also Meek 1971). And Adam Smith, consistently with this Scottish background, seems to present his version of it. On December 24, 1762, Smith tells his students “there are four distinct stages which mankind passes thro: —1st, the Age of Hunters, 2dly, the Age of Shepherds; 3dly, the Age of Agriculture; and 4thly, the Age of Commerce” (LJA i.27). During the semester he elaborates on this “natural progress which men make in society” (LJA iv.19). Smith explains that we start from hunting societies, where the “the first method [these small societies] would fall upon for their subsistence would be to support themselves by the wild fruits and wild animalls which the country afforded” (LJA i.27). By eventually realizing that domesticating animals provides an easier source of food, we then arrive at the age of shepherds. Population growth would follow a more stable supply of food and would push for an ever more steady provision. “Then they would naturally turn themselves to the cultivation of land and raising plants and trees as produced nourishment fit for them…. And by this means they would gradually advance into the age of agriculture” (LJA i.31). So, after having enough flocks and herds to be able to afford to cultivate some land, and after having enough produce to support themselves, people will have superfluous products with which to start to trade. And society will eventually develop into its age of commerce. Each stage of development has its appropriate form of justice and of government and its appropriate customs. And from each stage allegedly a society will eventually carry into the following stage. This is “the normal process of development” (Skinner 1975, p. 155). The model seems simple, clear, and quite reasonable.

Yet, if I were to give an historical example to explain and corroborate this model, I would say: take country x for example. During century y, people were few and they were hunters. As population increased, they became shepherds. In century yy, more or less, they started to cultivate land extensively and, now, x is in its commercial age. But this is not what Smith does. This is not the way Smith gives empirical support to his model. He tells his students that North American Indians are an example of the first stage of development, “tho they have no conception of flocks and herds, have nevertheless some notion of agriculture … [even if] this can hardly be called agriculture” (LJA i.29). “The Tartars and Arabians subsist almost entirely by their flocks and herds. The Arabs have little agriculture, but the Tartars none at all” (LJA i.29). And France is offered as an example of commercial society, without any previous stage of it being mentioned. We do encounter the Tartars and the Arabs again, on “Febry 21” and “Febry 23,” 1762. But here Smith tells his students that this natural gradual development somehow stalled with them as they are “two great nations who have been merely shepherds as far back as we can trace them and still are so without the least agriculture” (LJA iv.36).

Even if we try to think of Smith as using conjectural history—conjecturing how societies might have developed in the absence of direct evidence by looking at different societies—so that “at the beginning we were all America,” we run into the problem that we should all have been Tartary too, and eventually we will all become France. But, here again, this is not what Smith tells us. He tells us that America did not become Tartary, as we will see below; that the Tartars have always been and always will be Tartary, because of the nature of their land, as mentioned above. And France, well, we are simply not told. France is just France. These may be good examples of classification, not of development.

So, if we believe that Smith uses the stadial model of development as a model of development, then Smith tells his students there is a model of economic development and offers them examples of societies that do not develop. The situation does not improve when we look at what Smith tells posterity with his published works.

III. WEALTH OF NATIONS

In the Wealth of Nations (WN) the explicit mention of the four stages of development disappears. There are references to hunters, shepherds, agricultural, and commercial societies, and to development, especially in books III and V, but there is no explicit use of a model of linear development from one stage to another. Yet, if we really want to keep the model of the four stages of development in mind, we run into additional empirical problems.

Assuming the implicit presence of the four stages model, what does Smith tell us of civilizations that have evolved over time? Take Rome. Smith tells us “Rome … was originally founded upon an Agrarian law” (WN IV.vii.a.3, p. 556). Where is the age of shepherds, not to speak of the age of hunters, that precedes the age of agriculture? They may have been there, but this is not what Smith tells us. Rome was founded on agrarian law—no mention of what was there before, if anything.

Let’s look at something more recent, then. Book III is a broad account of the feudal era, from its beginning to its fall. Here, Smith also tells us that “the cultivation and improvement of the country, therefore which affords subsistence, must necessarily, be prior to the increase of the towns, which furnishes only the means of conveniency and luxury” (WN III.i.2, p. 377). He also tells us, “According to the natural course of things, therefore, the greater part of the capital of every growing society is, first, directed to agriculture, afterwards to manufactures, and last of all to foreign commerce. This order of things is so very natural, that in every society that had any territory, it has always, I believe, been in some degree observed” (WN III.i.8, p. 380). This seems to echo his lectures: first one develops agriculture (the country) and then, with the surplus products, one starts trading and developing trading centers such as cities.

This is a good place to argue that Smith adopts the theoretical model of stages of development. The chapter where this idea is presented is indeed titled “Of the Natural Progress of Opulence.” But Smith describes the development of Europe as actually “unnatural and retrograde” (WN III.i.9, p. 380). Rather than going from the agricultural stage to the commercial one, Smith tells us, we went from an agricultural stage back to a barbaric stage, to then jump into a commercial stage, following which agriculture improved. Europe developed foreign trade first, then manufacturers, and only later agriculture. This is not a minor point Smith makes. He spends three of the four chapters of Book III telling his readers the story of how the natural order of things is inverted. He even explicitly warns his readers of this inversion of the natural course of things at the end of the first chapter by saying: “But though this natural order of things must have taken place in some degree in every society, it has, in all modern states of Europe, been, in many respects, entirely inverted” (WN III.i. 9, p. 380).

A combination of exogenous events such as the barbaric invasions and of local politics, ranging from primogeniture laws to privileges the kings would grant to cities in exchange for support against the nobility, gave incentives to people to cluster in cities for protection, to develop trade, then manufacturers, and only later on to export to the countryside the entrepreneurial spirit of commerce that develops agriculture, rather than vice versa. This is why the European development had been “unnatural and retrograde.” The historical data we are offered give us exactly the opposite result from what we expected. The Meek-like theoretical model, therefore, by itself, has problems in explaining historical facts (note that Meek (1976, pp. 228–238) himself believed that the four stages theory was conjectural).

Robert Blecker (1997, p. 534) explains the relevance of this “unnatural and retrograde order”:

It allows for the endogenous development of absolute advantages and surplus capacities by exploiting scale economies through the market-widening effects of trade. … It is a view of economic development as a process in which countries develop in an interdependent manner, rather than autonomously. In the “unnatural and retrograde order”, the evolution of national economies is fundamentally influenced and determined by their position vis-à-vis each other in the international division of labour. In this vision of an integrated global economy, an international division of labour arises endogenously along with the productive capabilities of the nations that comprise it, as nations tend to develop absolute advantages cum surplus capacities in those goods in which (for reasons that may be historical accidents) they initially specialize.

Should Blecker be correct, the question remains, and becomes even more resonant: Why would Smith offer us a model of “natural development of opulence” if his actual model is the “inverted” one?

Let’s consider another factor. In Lectures on Jurisprudence, Smith tells his students that the engine of change is population growth. One society develops enough means of subsistence to sustain a larger population and, because of the larger population, is pushed to find additional means of subsistence, going from hunting to the more stable domestication of animals, to agriculture and finally to commerce. This feedback loop is also presented in Wealth of Nations, even if the emphasis is now more on population growth as a symptom, not a cause, of economic development: if there is more to survive with, more people will survive (WN I.viii).

What, then, does a Meek-like model of development as Smith allegedly presents it tell us? Not much, it seems, when the historical account of it is taken into consideration. Customs, institutions, and other historical accidents seem to play a larger explanatory role than the model itself.

In defense of Smith, it is often claimed that, for Smith, the American experience resembles his theory (WN II.v.21, p. 366; and III.i.5, pp. 378–379). But if we take Anthony Brewer (1998) seriously, even this is an empirical failure of Smith’s alleged theoretical model. Brewer claims the North American colonies cannot be considered an example of natural development for two reasons. First, the process of colonization itself implies that the barbaric colonies are colonized by civilized cultures. “An open economy surrounded by more civilized centers” (1998, p. 93) can hardly be considered the norm; rather, it is an exception. “Second, … [i]n the North American colonies, of course, modern institutions were imported from Europe, where they were … the result of a (much slower) process of development” (1998, p. 93). And, indeed, Smith tells us that the difference between the success of America and the degradation of Bengal is the adoption of the British constitution in America (WN I.viii.26, p. 91).

Looking at Book V of Wealth of Nations, where we have the most explicit reference to the four stages, still does not improve the performance of the model as a model of development. Smith uses the description of the four different states of societies to explain differences in military organization and performance. Techniques and discipline of the military change with each different state. The military develops from a group of thugs to a proper army. So far, so good. But then we have the description of Rome, both in Book V and in Book III. The sophisticated and civilized Roman Empire, given its proven powers, relaxed its military discipline (WN V.I.a.36) and barbaric, uncivilized German tribes overran it. “The rapine and violence which the barbarians exercised against the ancient inhabitants, interrupted the commerce between the towns and the country. The towns were deserted, and the country left uncultivated, and the Western provinces of Europe, which had enjoyed a considerable degree of opulence under the Roman empire, sunk into the lowest state of poverty and barbarism” (WN III.ii.1, p. 382). This is a step backward, a destruction of civilization (Skinner 1975, p. 159) that challenges the alleged inevitable linearity of the development of the four stages.

In Smith’s defense, one could claim that he tells us that the real commercial societies are in fact superior to any societies in their previous stages of development. Only opulent, therefore only commercial, societies can afford the expenses of modern firearms. Yet, he also tells us that “the unavoidable effects of the natural progress of improvement have, in this respect, been a good deal enhanced by a great revolution in the art of war, to which a mere accident, the invention of gun-power, seems to have given occasion” (WN V.1.a.44, p. 708; emphasis added). Poor and barbarous nations overrun opulent and civilized nations, unless we introduce firearms. But firearms are introduced by “mere accident.” The model, if it works at all, works only with an exogenous random event.

IV. CONCLUSION

A book inquiring into the nature and causes of the wealth of nations ought to be a book on economic development. It is, as it is many other things. There, as well as in his Lectures, Smith seems to present models of development, and at the same time offers a long range of historical examples with which to understand the models and their explanatory powers. Interestingly enough, though, the examples he chooses to use do not fit the stadial model of development. They all contradict it.

Smith seems to tell us we want to model the economic system linked to the progress of opulence, which took centuries and centuries to develop. But he also seems to tell us opulence developed in unpredictable ways. It happened only in a certain part of the world, and even there not in a stable way and not necessarily in its most effective form. A system that connects in our imagination the different points of history may be no different from a system that connects in our imagination the different points of the skies. It may be a product of our imagination that needs to be changed under the pressure of exceptions and contradicting evidence.

If Smith does use the four stages of development, not just as a taxonomy of different kinds of societies but as an actual model of development, as Meek and Meek-like interpretations do, why does Smith offer so much contradicting data? I would not go as far as Brewer (1998, p. 93) in saying “in sum, Book Three of the Wealth of Nations is a bit of fraud.” Yet, Smith’s results of his “empirical” tests are indeed problematic and may open the door for questioning the nature of the model itself. For Smith, is the deterministic model of stadial development commonly adopted by his peers actually the appropriate model to explain development? What if a better model is the “inverted” model (for a recent model of cities as economic drivers for the country, see Duranton and Puga 2019)? Is Smith hinting we should question the presence of a natural order in nature that can be modeled, or question our ability to do so? What if the system of natural liberty that is part of the economic development is indeed the result of peculiar circumstances and historical accidents? What if freedom and prosperity are generated simply by good luck (Forbes 1975, p. 198)? Smith does not seem to be able (or willing?) to exclude that possibility.

If we do not consider the stages of development as a simple taxonomy, and if we want to save the possibility of a stadial developmental as a model of development, what Smith seems to indicate with his empirical examples is the inappropriateness of the kind of model used. When the empirical data do not fit our model, maybe it is time to inquire again.