As de Ste Croix says, the fact that few members of the ancient Greek populations were either slaves or slave-owners is typical of economies where chattel slavery is a legal form of labour, and not that relevant, because the crucial fact is that the majority of the population were -- unlike the slaves and slave owners --- not usually involved in surplus extraction. The following is a poor subsititue for de Ste Croix himself, but it'll do for now:
"An exchange on G. E. M de Ste. Croix, historian of Ancient Greek society" By Ann Talbot 8 April 2000
"[D]oes de Ste. Croix think the Ancient Greek city-state was divided into propertied and unpropertied? Yes, and so does Aristotle. This is not, however, the same as saying that most of the people in a Greek city-state were either rich or poor. Most of them fell somewhere in between. They were what Aristotle calls the hoi mesoi, the people in the middle, whom he discusses in book IV of the Politics. He explains that the best state has a large number of these people of moderate wealth, because they will support neither an extreme oligarchy nor an extreme democracy. But as de Ste. Croix points out:
"On the other hand, Aristotle also (and more often) resorts to a simpler 'dichotomic' model-which, by the way, is regularly adopted by Plato. In Aristotle's dichotomy (as in Plato's and everyone else's) the citizens are divided into rich and poor, or into the propertied class ( hoi tas ousias echontes) and those who have no property, or virtually none ( hoi aporoi). Even in the passage from Politics IV that I have summarised above Aristotle admits that the number of mesoi in most cities is small, and he regards outright oligarchy or democracy as only too likely to occur. In general, it would be true to say that in Aristotle, as in other Greek writers (especially the historians), the nearer a political situation comes to a crisis the more likely we are to be presented with just two sides: whatever the terminology used (and the Greek political vocabulary was exceptionally rich) we shall usually be justified in translating whatever expressions we find by the 'upper classes' and the 'lower classes' meaning essentially the propertied and the non-propertied" ( The Class Struggle in the Ancient Greek World, p. 72 ).
He comments:
"The most important single dividing line which we can draw between different groups of freemen in the Greek world is, in my opinion, that which separated off from the common herd those I am calling 'the propertied class'..." (p. 114).
I hope I've said enough to show that de Ste. Croix thinks that the essential division in the Greek city-states was between rich and poor and that he backs this up from the sources, although by no means all, or even most, ancient historians would agree. More commonly historians emphasise, as you do, that the majority of the citizens were Aristotle's hoi mesoi, peasants who worked the land with the help of their families and possibly a few slaves.
This arithmetical approach seems to be common sense and even materialist. We ask what the majority of the population do for a living, what their relationship is to the productive forces and find in the case of ancient Greece that they were peasants. So we declare that this is what determines the character and dynamic of this society. This is a purely formal approach to the problem, however. Marx comments on this method in the Critique, where he writes:
"When examining a given country from the standpoint of political economy, we begin with its population, the division of the population into classes, town and country, the sea, the different branches of production, export and import, annual production and consumption, prices, etc.
"It would seem to be the proper thing to start with the real and concrete elements, with the actual preconditions, e.g., to start in the sphere of economy with population, which forms the basis and subject of the whole social process of production. Closer consideration shows, however, that this is wrong. Population is an abstraction if, for instance, one disregards the classes of which it is composed. These classes in turn remain empty terms if one does not know the factors on which they depend, e.g. wage-labour, capital, and so on" ( Contribution to the Critique of Political Economy, Progress Publishers, Moscow, 1970, p. 205).
In the course of history there have been many thousands of societies in which peasants have made up the majority of the population. For example, in twentieth century Latin America, Asia, Africa and large parts of Eastern Europe most people have been peasants. Yet these societies are very different from those of the ancient world. The designation "peasant", that appears at first so concrete, is, as Marx puts it, an "empty term", which does not help us to distinguish between the many societies in which the majority of the population have made their living by small-scale subsistence agriculture. Nor does it help us to identify what inner contradictions provided the dynamic of any of those societies.
By contrast, Aristotle, that consummate dialectician of the ancient world, has identified one of the fundamental contradictions in the Greek city-state, which is the key to the laws of motion in that society. When he says that the nearer the political situation in a city comes to crisis, the more apparent the division is between rich and poor, he is observing a general dialectical law at work.
The fact that few societies in which the majority of the population are peasants have achieved the level of cultural and political development that existed in Ancient Greece, should tell us that it is not sufficient simply to say that the majority of the population are peasants. We have to understand the essential historical character of a society if we are to invest a term like "peasant" with any meaning. We cannot know that twentieth century peasant societies are determined by their relationship to the capitalist world market by counting heads, but only by an examination of the historical development of these societies, which reveals their colonial and semi-colonial relationship to the imperialist powers.
In the case of Ancient Greece the surplus that paid for the monuments and allowed a small elite the leisure to write plays and poetry, and develop philosophy-and a rather larger group the leisure to watch plays, listen to poets and philosophers or attend the assembly-was produced by slaves. As de Ste. Croix himself puts it, " the most significant distinguishing feature of each social formation, each 'mode of production', is not so much how the bulk of the labour production is done, as how the dominant propertied classes, controlling the conditions of production ensure the extraction of the surplus which makes their own leisured existence possible." (p. 52, emphasis in the original).
This exploitative relationship between citizens and slaves is the other great contradiction in ancient society, apart from that between rich and poor. Or perhaps it is better to describe them as aspects of the same fundamental contradiction. The ferocity of the laws pertaining to slaves and the fear, acknowledged in Greek and Latin literature, that every slave was an enemy harboured, amply testifies to the centrality of this contradiction to the ancient world.
One writer who rejects this analysis is Ellen Meiksins-Wood, a radical historian connected with the journal Monthly Review. In her Peasant-Citizen and Slave: the Foundations of Athenian Democracy (Verso, 1988) she argues, "the distinctive character of Athenian democracy was not the degree to which it was based on dependent labour, the labour of slaves, but on the contrary, the extent to which it excluded dependence from the sphere of production, that is, the extent to which production rested on free, independent labour" (Meiksins-Wood, p. 82). She accuses Marx and Engels of adopting what she terms the "myth of the idle mob" and of inventing something called "the slave mode of production", a concept, by the way, which is not to be found in Marx. She takes de Ste. Croix to task for giving a large estimate of the number of slaves in Ancient Greece, as though the role of slavery in a society was a purely quantitative question and there must be a majority of them before we can say that slavery was essential to ancient democracy.
The constitutional measures, associated with the names of Solon and Kleisthenes in the sixth century BC, which established democracy in Athens, produced a qualitative change in Ancient Greek society. They prevented the propertied class from exploiting the peasantry as they wished, with the result that they increased their exploitation of those who could not defend themselves-the slaves.
We can only arrive at this conclusion if we analyse Ancient Greek society as an organic system, in which we are alert to its living processes and do not simply try to pin down its structures like beautiful butterflies in a museum."
http://www.wsws.org/articles/2000/apr2000/croi-a08.shtml