The way in which a given society organizes its productive processes is a crucial variable in our understanding of its wider socio-economic structures. For historical societies, the nature of such can be uncovered by analysing the types of labour apparent, and especially by assessing which ones were dominant. Of all types of labour relation, work referred to as wage labour is of particular importance in the study of ancient economies, insofar as its existence is an accepted indicator of market-based economic activity; indeed, in formalist approaches, the presence of wage labour is used as a yardstick for a given economy’s development, while substantivist thinkers see the emergence of wage labour as a crucial driver behind the advent of the market.

This chapter offers an overview of some of the theories underlying the connection between wage labour and market society. As we shall see, often definitions of what exactly constitutes wage labour are not always clear-cut – ranging from the simple provision of payment to the need for a whole set of conditions under which the work is to be carried out. Naturally, these different academic attitudes to wage labour have led to diverging identifications of this type of labour in the historical sources, and they underlie the – often contrasting – interpretations of its existence, or indeed frequency and scale, in debates on the nature and structure of the ancient economy. This study contends that, within the ancient world, the first large-scale instance of wage labour can be traced in the military sphere, where, from the late Classical period onwards, a growing number of soldiers served under conditions that align with those of the wage labourer, the conditions for which are assessed below.

2.1 Labour and Economic Structure

The history of work and the analysis of the types of labour apparent are established themes in studies on the wider social, political, and economic arrangements of historical societies.Footnote ¹ After all, the organization of the productive process not only sheds light on a given society’s resources and allocations, but also on the ways in which the required tasks were completed, by whom, and under what circumstances: the latter conditions are societally determined and in turn influence the structure of the communities in which they occur. These conditions, which are referred to as labour relations,Footnote ² have featured heavily in the grand narratives of history, whether penned by sociologists, economists, or historians. Weber famously saw a link between the Protestant ethic and the eventual development of capitalism, and regularly drew on labour history in order to explain general historical phenomena.Footnote ³ Earlier, Adam Smith, who saw labour as the true value of all things, also analysed the stages of historical development in accordance with the division of labour apparent.Footnote ⁴ The clearest expression of changes in labour as a relevant variable in historical analysis is found in Karl Marx’s concept of the mode of production (MoP), which refers to and brings together the context and conditions under which work is carried out. The resulting mode of production shaped wider societal frameworks: different modes of production therefore resulted in different types of society.Footnote ⁵

Marx’s concept added some analytical prowess to the various stages of development envisaged by Smith. However, although this approach has been hugely influential, it has rightly received its fair share of criticism, too. The model is considered as somewhat underdeveloped, as is especially clear from its tendency to categorize historical societies and their perceived organization of the productive process into preconceived moulds. For instance, the most prominent modes of production are the sequential modes labelled as slave, feudal, and capitalist; these exist in parallel to the European epochs of Antiquity, the Middle Ages, and modernity. This immediately reveals the model’s Eurocentrism, the problems of which are brought out most clearly in its assessment of Near Eastern societies, which were deemed to have operated under the so-called ‘Asiatic mode of production’, and which have been duly criticized for associating non-western societies with despotism and social stagnation.Footnote ⁶

Nonetheless, while problems with the concept of the mode of production can easily be found, there are benefits to the approach outlined. Thus, Eric Wolf, who endeavours to avoid needless over-categorization by envisaging more generalized modes that are described as kinship, tribal, and market modes,Footnote ⁷ advocates for the benefits that close analysis of a society’s productive process can bring to comparative studies. Simultaneously, he emphasizes the communal nature of all work, whereby studies of economic processes can more easily be related to social aspects.Footnote ⁸

Recent developments in theorizing labour history likewise focus on improving the descriptions and subsequent categorizations of different types of labour, and call for a specifically ‘global’ labour history that explicitly includes non-western and pre-industrial societies.Footnote ⁹ In doing so, this approach aims to expand the focus of labour history beyond the last two centuries that have long tended to command labour history’s focus, while the expansion of the chronological scope also aids in avoiding the framework of the nation state.Footnote ¹⁰ In offering taxonomies of labour types, the focus lies on changes in labour relations and the factors contributing to these.Footnote ¹¹

Nonetheless, all studies of work and labour – whether conducted by historians, sociologists, or economists – should be approached with their broader theoretical context in mind. As outlined by Chris and Charles Tilly, three main strands dominate theories of labour: the neo-classical, the Marxist, and the institutional approaches.Footnote ¹² The neo-classical view is inevitably based on the notion of the universality, in both space and time, of economic behaviour; the theory’s main agent, the so-called Homo economicus, is forever driven by profit-maximization behaviour and makes rational economic decisions towards the fulfilment of this aim. The neo-classical approach to the study of labour is therefore built on the same principles as those it holds of the economy more generally, seeing labour as a transaction aimed at maximization of profit and as something carried out by individuals. Apparent differences in labour relations, such as the differences between serfs and slaves, are recognized but seen as transitory.Footnote ¹³ Economic change is viewed as caused by technological developments and population growth.Footnote ¹⁴

Understandably, doubt has been cast on the presumed universality of human profit-maximization and rational behaviour that underlies this neo-classical approach.Footnote ¹⁵ Especially within the field of labour, a number of constraints arise that might impede it: not only may individuals have varying priorities, but the rationality of decisions made without access to all the information is always questionable.Footnote ¹⁶ Both the Marxist and institutional approaches differ from the neo-classical perspective in so far as they allow for society to play a fuller role in the structuring and organization of labour; here, class and customs, respectively, inform views and, consequently, approaches to the study of the labour process.Footnote ¹⁷

Despite such differences, all three approaches concur regarding the connection between labour types and the development of market society. In particular, the existence of the wage labour relationship is taken as indicative of the presence of a market more generally: those receiving wages will need to rely on the market to acquire necessities; the search for and subsequent competition over workers to produce these goods and services gives rise to a labour market.Footnote ¹⁸ The association of the wage labour relationship and the market society – often dubbed ‘capitalism’ – might explain why labour studies have so long focused on the era following the Industrial Revolution, when sweeping changes to labour relations fundamentally changed the productive process, and resulted in growing urban markets, higher urban demand, and subsequently increasing numbers of workers working for wages.Footnote ¹⁹

Simultaneously, this connection also means that evidence of wage labour in pre-industrial society can be taken to imply the existence of market-based economies. Indeed, within the formalist–substantivist debate, there is consensus on the importance of wage labour to the identification of market societies in history. As we shall see, for formalists, the presence of paid work is often perceived as evidence for the existence of a (developed) market, and accordingly as evidence for market society. Substantivists, on the other hand, while agreeing on the fundamental connection between wage labour and market-based economic activity, less readily accept all instances of paid work as wage labour. Moreover, while formalists consider the emergence of wage labour as a natural development of and conclusion to the increasing division of labour, substantivists see the emergence of wage labour as a rupture, the creation of which gave rise to the market in the first place.

The concomitant development of wage labour and the market is fundamental to Polanyi’s substantivist view of socio-economic transformation. In his view, the economic laws identified by neo-classical economics can apply only to modernity’s market societies.Footnote ²⁰ The economies of earlier societies, by contrast, need a different type of analysis, focused on the social ties and institutions which governed contemporary economic behaviour.Footnote ²¹ He thus argues for a turning-point in time in which the economic structure as described by the substantivists gives way to that of the formalists. Unlike Smith and Marx, Polanyi does not see the emergence of the market society as a natural conclusion to a natural transition, caused by technological developments or class struggles, but rather as the result of state intervention.Footnote ²² The commodification of labour – and thus the emergence of wage labour – he argues, occurred as people were driven from working their own land, and were forced to compete with each other in the sale of their labour to acquire the necessities of life in an increasingly urban world.Footnote ²³

The model that relates wage labour to the advent of market society has been used in connection with the Industrial Revolution, which is seen to derive from the changing nature of labour relations in the sixteenth century, when the Enclosure Movement in England turned much common land into private property.Footnote ²⁴ As long as there was common land, the argument goes, all people had access to it in exchange for tribute; production of other necessities took place within individual households and in a village context. Markets existed in both town and country, but were peripheral, operating in accordance with different rules that protected against speculation, and dominated by a ‘moral economy’, according to which the price of a given commodity was determined by extra-economic forces and not aimed directly at achieving a profit.Footnote ²⁵ The goods sold on such markets, furthermore, were not produced for that purpose, but rather constituted the surplus of goods produced for and within the household. In this regard, the feudal system had allowed labour to provide workers with their daily necessities, and time spent working was ‘rewarded’ with the goods required, of which agricultural produce was the most important. The privatization of the common land made ‘commercial landlords’, ‘capitalist tenants’, and the ‘hired wage labourer’ the key players of the newly emerging economic structure.Footnote ²⁶ As argued by Meiksins Wood,Footnote ²⁷ this restriction of access to land, and thus to the traditional means of the production of food, led to competition for this land, and those lucky enough to hold a portion, namely the landowners, rented out the usufruct of this land to tenant-farmers, who in turn employed farmers with no usufruct of land to work it. The bulk of the produce generated on this land came to be sold on the market rather than immediately consumed. The element of competition for access to and subsequently the retention of land incentivized improved production for profit, giving rise to further divisions of labour, from which modern market society grew. Hence, the serfs of a feudal society became the wage labourers in an economy increasingly characterized by market exchange.Footnote ²⁸

Wage labour is therefore deemed a critical variable in assessments of the historical existence of what we call ‘the market’, and indeed of market society. Debate, however, centres on what exactly constitutes wage labour, and, as we shall see, its identification in the source record is dependent on the definitions applied.

2.2 Defining the Wage Labour Relationship

Despite agreement on the importance of wage labour to economic development and change, definitions of exactly what type of work constitutes wage labour are more difficult to come by. Because of its link to market society, wage labour has become tied up with discussions on the historical development of the latter, its historical existence taken as evidence for the universality (or lack thereof) of the market and market-oriented behaviour. Proponents of the latter view often opt to see the wage labour relationship in play whenever payment is made for services rendered, while others prefer a more specific set of characteristics surrounding the employment relationship to identify paid work as wage labour.Footnote ²⁹

As mentioned, the view of the wage labourer as a fundamental agent of economic change was first professed explicitly by Karl Marx in the mid-nineteenth century. Building on Adam Smith’s theory of the increasing division of labour, Marx took a fundamental conceptual step in the analysis of the economic relevance of labour: by drawing a distinction between labour and labour power, he introduced the principles according to which it could be analysed as a commodity – the reason for this type of labour’s special relation to market society.Footnote ³⁰ In this conceptualization, labour refers to the act of the production of goods and services in the most general form and therefore includes the raw materials, skills, and the act of work.Footnote ³¹ Labour power, on the other hand, refers solely to the human input in the productive process, and is defined as ‘the aggregate of those mental and physical capabilities existing in a human being, which he exercises whenever he produces a use-value (i.e. a good or service) of any description’.Footnote ³² In Marx’s view, the emergence of wage labour constitutes the commodification of labour power, which is expressed in time (i.e., work hours) and remunerated with (monetary) wages. Crucially, this is distinct from the fruits of the act of labour. A craftsman who sells a pot, for which he sourced the raw materials, acquired the tools for its production, and spent time making it, sells a final product which contains his labour power. When he sells his labour power, on the other hand, the act of working is being isolated from the other components included in the good or service; it is thus in itself treated as a commodity that is distinct from both the worker and the product.

Based on this conceptualization of the commodification of work, Marx established three conditions that allow for the purchase of labour power, and thus for the emergence of wage labour:

• the owner of the labour power must offer it for sale on the market;

• the labour power should only be sold for a definite period;

• the labourer should sell only their labour power, not commodities in which this labour power is incorporated.Footnote ³³

The first two of these conditions relate to the free status of the labourer; if the sale is conducted and the wages received by a third party, or if it is sold for an indefinite period, then the labourer’s status naturally reverts to that of a slave or serf. The third condition refers to the importance of the sale of labour power, rather than its fruits. However, Marx’s criteria for wage labour have led to much confusion and have not been immune from criticism, being described as unnecessarily restrictive and as exclusively applicable to the working classes of the nineteenth century.Footnote ³⁴ Instead, arguments have been made pointing to various forms of wage labour that would incorporate the wide variety of work performed in exchange for pay. One recent attempt at a broad typology of wage labour draws a distinction between the ‘carriers’ and ‘possessors’ of labour power, which are labelled as ‘autonomous’ and ‘heteronomous’ wage labour respectively.Footnote ³⁵ From this, the types of wage labour shown in Table 2.1 have been established.Footnote ³⁶

As displayed in Table 2.1, the distinction between autonomous and heteronomous wage labour allows for a much broader range of labour activity to fall under the umbrella term of wage labour – in fact, any work for which any form of remuneration is given can be included. Hence, the typical wage labourer as described above is here characterized as the ‘free wage labourer’: as both carriers and possessors of their own labour power, these individuals can exploit their labour power as they please. Effectively, they are distinct from, for instance, the sub-contracted wage labourer, whose labour power has already been bought and is merely being resold. In the context of the ancient world, the inclusion of slavery within this typology of wage labour is especially important. Chattel slaves are included as neither owner nor sellers of their own labour power – the latter would be sold by the slave’s owner, who could collect and keep any proceeds if he chose. In addition, the typology accounts for those slaves who, while not owning their own labour power, were able to sell it; that is to say, such slaves were able to work for wages directly accorded to them. Slaves receiving pay are indeed attested in the ancient sources, and in fact seem to constitute the largest pool of specialist workers in long-term employment, a famous Athenian example being the public slaves tasked with checking and approving silver coinage.Footnote ³⁷ Thus, by allowing for differences in ownership of labour power, this typology allows for a much wider interpretation of what constitutes wage labour, and by extension reveals a much higher incidence of wage labour among the total workforce.

However, within this broad conceptualization, the sole characteristic of wage labour is the provision of pay, while the wage labourer’s aforementioned free status, and concomitant ability to determine the price at which he chooses to sell his labour power, is of no relevance. This, of course, raises the question of whether payment for any given activity alone constitutes the existence of wage labour proper. We have previously seen the example of the craftsman who sells a product that contains his labour power, in contrast to the wage labourer who only has his labour power to sell and does so in exchange for a wage. The pay accorded to either worker is therefore to be recognized as being of a different nature.

Such distinctions should be borne in mind when discussing all types of pay offered to individuals in exchange for a service performed. In the context of Classical Athens, for instance, any given citizen would have received daily payment for democratic participation.Footnote ³⁸ Although this payment is sometimes translated as a wage, realistically the citizens fulfilling their democratic duty cannot be described as wage labourers.Footnote ³⁹ Two main reasons can be cited. First, the flat-rate payments for democratic participation are said to have been introduced in order to make attendance at meetings possible for all.Footnote ⁴⁰ Second, such participation did not constitute a continuous source of employment.Footnote ⁴¹ The payments offered should therefore be seen as reimbursement of time and productivity lost elsewhere, though the amounts allocated were likely intended to cover only the bare necessities for the day.Footnote ⁴² It ought also to be noted that the amounts allocated for assembly attendance showed little variation over time and place, and thus constituted a somewhat arbitrary figure: remuneration did not increase when assembly attendance was low, nor decrease when an unexpectedly large number of citizens were willing to perform their political duties.Footnote ⁴³ Similar concerns can be raised in relation to the payments received by slaves: because of their unfree status, not all could opt to hire out their labour power, while those who were allowed to do so by their owners would not only have their bargaining power curtailed by their status, but were also subjected to pay their owners the apophora (ἀποφορά) – a significant share of their income.Footnote ⁴⁴

The tendency to conceive of all payments to individuals as wages can in part be explained by contemporary terminology – as we shall see below, in the Archaic and Classical periods, all such payments to individuals were referred to as a misthos (μισθός), which is often indiscriminately translated as ‘wage’.Footnote ⁴⁵ Seemingly, the fact that many of these payments were made in coin produces further unhelpful parallels to modern notions of wage labour. In other words, if the skilled slave was paid in kind, would their activity still be seen as wage labour?

Clarity on what exactly constitutes a wage – namely the payment rendered to a worker in exchange for labour power – is especially important when thinking about wage labour as indicative of market society and its development. For when the price of labour power is arbitrary and therefore not set on a market, it also limits the development of price-setting mechanisms elsewhere. In effect, when there are no increases or decreases in the price of labour power, and thus in the level of wages received, fluctuations in other price-setting markets are likewise limited. Payments offered in the absence of a market for labour power and on an infrequent basis should therefore be seen as reimbursement for productivity lost elsewhere, rather than as remuneration for productivity gained. It is this mechanism that commands the inclusion of the labourer’s (continued) free status in the definition of wage labour, since it allows the labourer to influence the price at which their wage labour is sold. For these reasons, the insistence on the free status of the labourer is of crucial importance in any analysis of the economic impact of the emergence of wage labour.

Accordingly, when speaking of wage labourers in the following analysis, I propose the following criteria for the identification of wage labour:

• the labourer needs to voluntarily hire out their labour power for a defined period;

• the labour power should be remunerated;

• the labourer can agree to the price of the labour power they offer for hire;

• the labour power is offered for sale on the market.

These conditions – namely the continued free status of the labourer and their ability to choose the price of their labour power – means that the price of wages is set on the market, rather than arbitrarily.

2.3 Wage Labour in the Ancient Economy

The central role of wage labour in the emergence of market society has, of course, accorded it a central role in the long-standing debate on the nature of the ancient economy. And yet, although scholarship readily acknowledges its importance as a key variable, studies on the topic are often plagued by the sharp dichotomy that exists along the lines of the noted substantivist and formalist approaches. Thus, two strands of scholarship are evident, with one side emphatically denying the existence of wage labour in Antiquity, and the other rather uncritically assuming its presence and prevalence, often simply because of a payment, rather than the varied set of social relations and conditions that would render ordinary work into wage labour.

In substantivist views of the ancient economy as fully embedded in wider societal structures, there is no room for any form of capitalist behaviour, and wage labour consequently becomes a theoretical impossibility. Thus, influenced by both Weber and Polanyi,Footnote ⁴⁶ Finley categorically denied the existence of wage labour, describing it as a ‘sophisticated latecomer’Footnote ⁴⁷ whose existence would have required a leap in conceptual thinking – namely the commodification of labour power – that the ancients simply did not manage to take.Footnote ⁴⁸ Instead, in the substantivist view as professed by Finley, the whole of the ancient world should be characterized as a slave society in which goods and services were produced either within the household or through the labour of slaves.Footnote ⁴⁹ The view of the ancient economy as slave-oriented exists in more recent scholarship on later periods, too. For instance, in a discussion (admittedly brief) of labour in the Hellenistic economies, Descat maintains that the period ‘did not experience profound innovation in the matter of labour’,Footnote ⁵⁰ and he contends that the use of slaves became more commonplace, since it was the most rational development imaginable.Footnote ⁵¹ Economic reasons aside, Antiquity’s alleged reliance on slave labour is often explained from a cultural angle. Citizens, it is sometimes said, abhorred hired labour;Footnote ⁵² among Athenians especially, such employment constituted dependency deemed akin to slavery, and was thereby incompatible with the fiction of the democratic egalitarian society.Footnote ⁵³ The refusal by citizens to engage in long-term employment therefore gave rise to the use of slaves in those sectors of the economy that required consistent engagement from its workforce.Footnote ⁵⁴

Nonetheless, despite such objections, it ought to be acknowledged that our surviving sources reveal plenty of instances indicating payment given for services rendered, which, for many scholars, constitute evidence that is incompatible with the substantivist view. Historians of Rome especially have argued for the existence of wage labour and of a clearly functioning labour market.Footnote ⁵⁵ Comparable arguments for the existence of wage labour in the Greek world have been made. These assertions often share a methodological ground, and they take the attestations of pay across a wide range of sources for a variety of occupations as evidence for the wage labour relationship. An exemplary instance of this approach is Loomis’ catalogue of attested misthoi in Athens, from which it is concluded not only that wage labour existed, but also that Classical Athens saw a labour market.Footnote ⁵⁶ These conclusions were recently endorsed by Harris and Lewis,Footnote ⁵⁷ who, while not explicitly arguing for wage labour, state that the variety of occupations attested clearly indicates increased (horizontal) specialization, and thereby an increasing division of labour.Footnote ⁵⁸

Yet neither the simple attestation of pay nor the existence of a variety of specialized occupations necessarily secures a presence of wage labour. For instance, Loomis’ catalogue essentially lumps together all forms of payment associated with activities that can be construed as labour – thus, instead of remuneration granted to hired workers in exchange for their labour power, we find payments accorded to public officers, to soldiers and sailors, or the fees of doctors, actors, artists, builders, and seers. These are not restricted to payments given for services rendered, but rather include a variety of costs, such as state payments for public service, or indeed travel allowances.Footnote ⁵⁹ While many of these may well include remuneration for workers’ labour power, they say little about the level of wages accorded in Classical Athens; civic payments, on the other hand, should perhaps be seen as reimbursement of sorts, rather than as pay for services rendered.

Such concerns regarding wages, and indeed ancient wage labour, were highlighted by de Ste. Croix: unlike most substantivist scholars, he argued that the concept of labour power was known in Antiquity, as reflected in ancient terminology applied to workers: the hired labourer is referred to as a misthōtos (μισθωτός), while the skilled craftsman is recorded chiefly as, for instance, a technitēs (τεχνίτης) or a banausos (βάναυσος).Footnote ⁶⁰ While both categories of worker may have received a misthos, they differed in so far as the misthōtos sold his labour power, while the skilled workers sold specific pieces of work, the price of which contained their labour power, alongside the raw materials and the tools of production.Footnote ⁶¹ Naturally, this distinction also draws attention to the need to distinguish between different types of payment, and to qualify each instance of a misthos being paid as linked to the sale of labour power.Footnote ⁶²

The validity of these concerns becomes clear when looking at the terms of employment which a fifth-century workforce may have enjoyed. For these, we can turn to the Erechtheum Building Accounts from Athens, which preserve details on both the types of workers involved and the payments granted.Footnote ⁶³ These accounts reveal great variety – and indeed complexity – in the terms of employment: senior workers, such as the ‘architect’ overseeing the project, were on long-term, annual contracts; other labourers were employed on day- or piece-rates. Noticeable in these accounts is how slaves and free men were working side by side, with pay for each being a drachma a day, regardless of status and position. While we therefore see a complex web of labour relations with a variety of workers of varied statuses and skills, it appears unlikely that their pay was determined via competition on a labour market. The building accounts effectively constitute an instance of hired labour, but do not necessarily indicate the long-term availability of wage labour, or indeed the presence of a competitive market.

By introducing these complications, de Ste. Croix came to the view that, while wage labour occasionally occurred in certain contexts, Classical Athens (and the ancient world more generally) remained an otherwise slave-based society within which ‘the propertied class derived its surplus mainly from unfree labour’.Footnote ⁶⁴ A similar view is expressed by Vlassopoulos, who postulates the independent household as the central unit of the economy, and sees wage labour as a source of income for the ‘impoverished proletariat’, whose services were bought by those who could not afford the costs of slaves.Footnote ⁶⁵ The sources in fact preserve some indication of a functioning labour market; for example, there are tantalizing references to a hiring market known as the kolōnos agoraios near the Temple of Hephaistos, where day labourers are assumed to have gathered in search of a day’s work.Footnote ⁶⁶ It is likely that such workers might normally find a job in highly labour-intensive sectors, such as agriculture, which would have had high demand for labour during the sowing and harvesting seasons; it is indeed in such contexts that many references to hired labourers are found.Footnote ⁶⁷

While we should therefore acknowledge the presence of wage labour in certain or specific contexts, its economic impact should not be overestimated, since it was generally scarce and infrequent or at best seasonal.Footnote ⁶⁸ This is relevant especially when discussing the presence of wage labour in relation to highly specialist occupations: even if the activities of, for example, doctors or poets fitted the descriptor of wage labour proper, it is unlikely they existed in sufficient numbers to have had any significant structural economic impact. Those wage labourers who may have operated in larger numbers, such as hired agricultural workers, would only have been in that role for short periods of time each year, nor should we assume constant and consistent demand for labourers from the building industry – in both cases, hired workers could, of course, have been slaves.

Despite reservations regarding the scale and frequency of ancient wage labour, de Ste. Croix rightly singled out the military as the one sphere of activity in which this type of labour can be seen to have developed in a meaningful way, briefly citing mercenary service as the first large-scale instance of hired labour.Footnote ⁶⁹ Similar considerations have recently been cited in relation to the Athenian navy.Footnote ⁷⁰ Nonetheless, overall studies of labour have often overlooked military service, perhaps because warfare is rarely seen as a productive activity, thereby falling outside the traditional scope of the historian of labour.Footnote ⁷¹ Nonetheless, the view of warfare as fundamentally destructive is increasingly disputed, and scholarship is ready to take stock of warfare’s (inadvertent) positive developments.Footnote ⁷² For instance, successful warfare can lead to surplus value for states and elites through territorial and economic advantages offered; similarly, single campaigns might lead to short-term gains in the form of booty.Footnote ⁷³ However, the productive capacity of warfare is ultimately irrelevant when debating the inclusion of military service into studies of labour: armies are built and organized in the same way as other industries – ultimately, they rely on the shared labour or commitment of individuals in pursuit of a larger goal.Footnote ⁷⁴

In terms of the study of the ancient economy, the significance of wage labour has been acknowledged but its meaning at times fundamentally misunderstood. The denial of the wage labour relationship’s existence in the ancient world, as professed by Finley and other substantivists, is in essence based on the dogma that the Greco-Roman world cannot be characterized as a market society; on this view, the emergence of wage labour becomes a theoretical impossibility, and therefore a marginal variable at best. At the other end of the spectrum, however, the attestation of simple (monetary) payments for services rendered is taken as enough evidence to argue for the widespread existence of the wage labour relationship. Yet such interpretations generally fail to consider the complex nature of the wage labour relationship and do not properly acknowledge that remuneration may be offered in different contexts.

2.4 Identifying the Military Wage Labourer

Although military service may not have featured regularly in histories of labour, the work and activities of soldiers can nonetheless be assessed within the parameters established to analyse and categorize labour relations. Individuals’ military service might be coerced, or socially expected, or indeed acquired by those in need of troops. For military service to constitute a form of wage labour, we therefore need to concentrate first and foremost on paid troops – usually, but not always, described in scholarship as ‘mercenary’ forces. Yet since not every payment immediately constitutes a wage, further conditions of service will need to be considered. As established above, for a worker to be identified as a wage labourer, they need not only to be remunerated for their time, but also to be and to remain of free status, so that they can enter a labour market. Methodologically, these criteria will be transposed to the military sphere by analysing the nature of soldiers’ initial enlistment and subsequent terms of service, and the types and extent of their remuneration, and by questioning whether the price of military labour was set on a market.

As stated, the primary condition for the identification of a wage labourer is that they are free. Of course, the bulk of ancient Greek military manpower was traditionally mobilized through conscription, and should thus be seen as a form of ‘coerced military labour’ insofar as the citizen soldier was expected to serve his polis on the potential penalty of social exclusion.Footnote ⁷⁵ For soldiers to qualify as ‘free military labourers’, on the other hand, they will have needed to enlist on a voluntary basis, as was the case for soldiers described as mercenary, yet not limited to those alone. Furthermore, this enlistment should have been of a temporary nature, in order to guarantee their continued free status, and to provide scope to (re-)enlist in an army of their own choosing. This condition necessarily entails that movement across military employers was possible and was deemed an acceptable course of action. These military labour relations will be discussed in Chapter 3, through an analysis of the nature of soldiers’ initial employment, which will draw a distinction between coerced and free military service, and of the ensuing terms of service, focusing specifically on the troops’ continued free status and mobility. In analysing whether soldiers enlisted voluntarily or not, the discussion will not be restricted to soldiery described as mercenary (and thus voluntary) by the sources but will rather look at recruitment using a variety of divisions; this will avoid unhelpful and potentially anachronistic categories that obfuscate the various types and conditions of service that existed in the late Classical and early Hellenistic periods.

The second condition for the soldiers to qualify as wage labourers is that they will have received remuneration for the sale of their labour power in the form of a wage, the price of which ought to have been determined on the market. Thus, for soldiers to satisfy the conditions of wage labourer in terms of the wages received, they need to have sold their labour power for a price that covered at least their daily maintenance requirements. To assess the soldiers’ remuneration, however, all possible types need to be considered, and the focus should not lie solely on cash payments known as misthoi. Indeed, it is commonly held that a state has three options for rewarding military labourers, namely by granting the possession of usufruct of land, by making cash payments or payments in moveable goods, or by granting rights, most notably the right to pillage.Footnote ⁷⁶ The analysis will also take into account initial incentives for enlistment, such as the provision of equipment or land, both of which were prerequisites for service in the citizen armies of the Classical Greek world. To assess soldiers’ remuneration, Chapter 4 will therefore take stock of the incentives offered to induce or facilitate enlistment; of the wages paid out in coin; and of any other additional benefits offered by the military employers.

While the validity of the conceptualization of the soldiers as wage labourers can be tested through an analysis of both the soldiers’ status as free military labourers and the nature of the wages provided to them, we also need to ascertain whether the level of these wages was set on the market. This question will be addressed in Chapter 5, which will look at market mechanisms within and across armies. In doing so, it will first assess the presence of an internal labour market. The latter is characterized by divergences in status and wages within a single army and is seen as functioning when career progress is possible through increased specialization, rather than via pedigree or other social institutions leading to, for instance, a position of command. The chapter will highlight the presence of varied pay scales in accordance with labourers’ skills, military labourers’ specialization, spurred on by incentives, and social mobility. By contrast, an external labour market is responsible for price-setting of wages more broadly, which is achieved through competition between various employers for labourers. Within the military context, its presence can therefore be detected through the attestation of the movement of labourers between employers, of collective action by free labourers, and by competition among employers as indicated by adjusting wages to attract labourers of diverse skills and abilities. In tandem, the chapter will explore less tangible indications of market mechanisms; in particular, it will illustrate that some soldiers engaged in what is termed profit-driven behaviour, as indicated by their demands for better terms of employment.

As I will argue in the following chapters, the development of military wage labour can be seen to begin with the military reforms of Philip II of Macedonia. From this point onwards, increasing numbers of soldiers fitted the descriptor of wage labourer, owing to their free status and the provision of remuneration. Nonetheless, while Philip’s reforms certainly put in place the all-important foundations, these developments did not come to full fruition until after the death of Alexander the Great, when his Successors waged war over and across his old empire, thereby competing over these free military labourers’ enlistment; the strongly martial character of later Hellenistic kings duly ensured the continued existence of a market for military labour in the Hellenistic period, thus giving rise to that key ingredient of economic development: the large-scale presence of wage labour.