From the Pre-Socratics to the Hellenistic Age (2024)

The search for Greek atheists

The year is 399 bce, and Socrates is on trial for his life. The charges include refusal to recognize the gods cultivated at Athens, and the introduction of new divinities. In Plato’s imaginative reconstruction of the trial Socrates, cross-examining his accuser Meletus, remarks:

Meletus opts for the latter alternative, imprudently turning the charge into one of outright (positive) atheism. And in Socrates’ quoted words we find the very first recorded occurrence, in any Western language, of a word for ‘atheist’. The adjective atheos had previously carried its literal meaning ‘godless’, but henceforth it came to serve in addition as a noun signifying one who upholds a specific creed: no gods exist. And it eventually acquired a cognate abstract noun, atheotēs, ‘atheism’.

A closely associated terminological point concerns another Greek word used by Socrates in the same passage, the verb nomizein, literally ‘believe’, and inadequately translated above as ‘recognize’. When ‘gods’ are its grammatical object, its semantic scope fails to distinguish between the outward practice of ‘cultivating’ gods and the inner state of ‘believing in’ them, that is, in their existence. The expression ‘not recognizing (the) gods’ was the favoured way of referring to what we today call atheism. However, the explicitly existential aspect, namely denial of the gods’ existence, is usually less emphasized than the cultic one: failure to take part in worship. The above passage is therefore unusual, to the extent that Plato has inserted an explicit mention of failure to recognize the gods’ very existence.

Not recognizing the gods was in principle a punishable offence in ancient Greece. It had, for example, been enshrined as such in Athenian law since the 430s bce, when the Decree of Diopeithes outlawed ‘not recognizing divine beings’ (ta theia mē nomizein). Indeed, around that time such a charge had reportedly been brought, under Athenian law, against the physicist Anaxagoras, for daring to say that the sun is not, as generally believed, a divinity, just a red-hot stone. The slur that so-and-so calls the sun a stone came to be emblematic of the charge of atheism. In fact Meletus’ very next move in the above exchange is to accuse Socrates of doing just that (26d).

Anaxagoras, despite his physicalist account of the sun, had in reality not been a full-blooded, positive atheist, nor had anyone else we can name from the period down to the end of the fifth century bce. Anaxagoras described a world whose creation was initiated by a cosmic ‘intelligence’ (nous), a power that has the hallmarks of a divinity, whether or not Anaxagoras himself so called it. From Thales (‘All things are full of gods’) in the early sixth century bce, to the death of Socrates at the beginning of the fourth, the default assumption of philosophers had always been that one or more of the divinities misconceived by Homer and other early poets as anthropomorphic beings are, properly reinterpreted, genuine causal forces in the universe (cf. Sedley 2007; Trépanier 2010).

How about Democritus, the founding father of atomism active in the late fifth and early fourth centuries, according to whom nothing exists but atoms and void? Instead of advocating atheism, as one might have predicted, Democritus posited gods made out of atoms, nebulous living beings capable of visiting us in our dreams, and of doing us good or harm. Why did he, of all people, not take that final step and eliminate the gods from his ontology? Part of the answer no doubt lies in the ubiquity of religious experiences, such as divine epiphanies in dreams, in a culture saturated in cultic practices, mythological narratives, and divine images. It may never even have occurred to him that these divine figures were in fact illusory. Another factor, witnessed by the Decree of Diopeithes, is the social and even legal opprobrium attached to a failure to cultivate the deities on whom the city’s well-being was supposed to depend. To come out as an atheist was not merely an intellectual decision, but an act of considerable courage.

Plato’s last work, the Laws, includes in book 10 an elaborate critique of (a) atheism, along with two other heretical positions, namely (b) that the gods exist but are uninterested in us, and (c) that, although they are interested in us, they can be bought off with sacrifices and other bribes. The Laws was written around 350 bce, but the dramatic date of the protracted conversation in which it consists is indeterminate, and could as well represent the late fifth century (as do most of Plato’s other dialogues) as the mid fourth. What the Laws makes clear is that, at this dramatic date, whatever it might be, atheism is rife in Athens. Plato’s main speaker, the Athenian stranger, remarks that, where he comes from, atheism is fashionable among the young, who rely on the authority of various written texts, some of them prose works, some verse.

Who were the authors of these texts? They may well have chosen to remain anonymous. At any rate, later writers who tried to identify early atheists had a hard time putting together a list of names. One favoured candidate was Prodicus of Ceos, a Sophist active in Athens in the late fifth century, who had suggested that the gods originated as deifications of commodities (e.g., corn, wine) and benefactors specially appreciated by early mankind. However, the attribution of atheism to him on this basis is likely to have originated as mere conjecture. Prodicus frequently appears in the writings of Xenophon and Plato, where he is treated respectfully, without the slightest hint that he was known as an atheist. Besides, the anthropological theory in question was certainly not enough to guarantee atheism, since it was taken up later even by the thoroughly theistic Stoics. It was after all possible to hold that those primitive deifications were early mankind’s recognitions of genuinely divine forces or beings. (However, for a lucid presentation and interpretation of the evidence for Prodicus that accepts the attribution of atheism, see Mayhew 2011.)

An older contemporary of Prodicus, likewise associated with atheism in the later tradition, was the great Sophist Protagoras. The evidence is quite clear, however, that rather than a positive atheist he was the first explicit agnostic. He opened his work On gods as follows:

We should nevertheless note in passing that, even if Protagoras was not a positive atheist, his words suggest a cultural context in which positive atheism was an established position, available to endorse or reject. Before someone can explicitly suspend judgement as to whether or not x exists, cases both for and against x’s existence are likely already to have some currency. Hence the Athenian atheist movement deplored by Plato in the Laws was probably in existence well before Protagoras’ death, c. 420 bce.

Another suggested early atheist was Diagoras of Melos (see Winiarczyk 1981), a controversial figure active in late fifth-century Athens, who was indeed known as ho atheos. However, it is now widely agreed that the epithet did not in this context mean ‘the atheist’, but ‘the godless’. For Diagoras had a price put on his head at Athens after notoriously mocking the religious mysteries.

The Sisyphus Fragment

The final name on the early list of supposed positive atheists is a more plausible candidate: Critias, an uncle of Plato, notorious for his membership of a vicious junta at Athens, the Thirty Tyrants, in 404 bce. He was reputed in some quarters (see Sextus Empiricus, Against the professors 9.54) to be the author of a scandalous dramatic passage, known today as the Sisyphus fragment, in which the speaker alleges that the gods were invented by politicians or lawmakers. Placed in the mouth of the mythical villain Sisyphus, the theory runs as follows.

Because the earliest laws proved ineffective against those who had found ways of committing crimes unobserved, someone had the brainwave of inventing superhuman supervisors who could see everything we do and even read our thoughts:

Locating them in the sky, Sisyphus continues, was itself a further stroke of genius, because the sky is at once the source both of the greatest benefits to mankind—sun and rain in particular—and of the greatest terrors, such as thunder and lightning. Thus the gods were at once a 24-hour satellite surveillance system, our greatest benefactors if pleased with us, and the greatest threat to our well-being if displeased.

This brilliant passage was undoubtedly one of the canonical atheistic verse texts invoked by the atheists according to Plato. Whether its author really was Critias is a harder question to answer. Other sources attributed it to Euripides, and this split in our evidence may well indicate that its true authorship was in reality unknown, a matter for conjecture alone. Given the social and at times even legal obstacles to public atheism, it would not in fact be all that surprising if the actual author, whoever he was, preferred the veil of anonymity. The same is likely to apply to the other atheist authorities mentioned by Plato.

The Atheists in Plato’s Laws

Whether or not they included Critias, the atheists criticized in Plato’s Laws had a well-developed position, which Plato takes the trouble to sketch to us. Contrary to a widely shared assumption on the part of scholars, there is no reason to think that Plato has invented this atheist case, piecing it together himself out of various fifth-century concepts and theories. Its coherence, and its inclusion of details which as far as we know are original, favour the assumption that there really was such a movement at Athens in the late fifth and/or early fourth century (see Sedley 2013). It amounts, if so, to the earliest comprehensive defence of positive atheism.

According to these anonymous early atheists (Laws 889a–890a), the world and its occupants originally came into being out of inanimate material elements, harmoniously shaped not by intelligence but by the interplay of natural forces. Human crafts provide an inferior supplement to, or imitation of, what nature has already achieved on a grand scale. Some of these crafts, for example medicine and agriculture, are founded in truth, because they build on nature’s workings, but others, including the creative arts, are basically fictive, and are therefore a poor guide to truth. Prominent among the latter, fictive crafts is legislation, which fabricates norms of justice and injustice: that justice and injustice are indeed mere human inventions, rather than founded in natural truth, is confirmed by the way their norms differ from place to place. Now what has been said about justice can equally be said about the gods: they too differ markedly from place to place and from culture to culture. This is because the gods have been instituted in each city by law, and are, along with justice, a mere legislative device. (It is in this final stage that we can see the Sisyphus fragment, with its attribution of the invention of the gods to a pioneering legislator, providing a vital component of the theory.)

This is a globally conceived theory, locating the invention of gods within a complete world history in order to demonstrate the point at which they entered: not at the outset, in the origination and structuring of the world, but at the very end, where human society was inventing means of stabilizing itself.

So ambitious an account could not have been constructed before the late fifth century, when for the first time it was possible to combine a theory of the world’s purely accidental origin, as developed by the atomists Leucippus and Democritus, with an anthropological theory—typical of the Sophistic era—concerning the origins of social institutions, law and religion included.

Plato, in his final years, devoted a remarkable amount of space to arguing and (hypothetically) legislating against atheism. What was at stake? Atheism is presented in the Laws as undermining morality, by reinforcing the call of those radical moralists who advocate an abandonment of ‘conventional’ or ‘legal’ values—the Greek noun nomos covers both convention and law—and a reversion to ‘natural’ norms whereby the strong may dominate the weak without restraint. We cannot know whether the atheists targeted in Laws 10 all drew this consequence, but it is likely enough that some did. Indeed, we can be confident that in the Sisyphus fragment, one of their canonical texts, the villainous Sisyphus (later punished by having to push a stone uphill in Hades for all eternity) was using his atheist declaration precisely in order to justify his own life of wrongdoing. Greek religion had evolved in partnership with law, and we should not be surprised if liberation from the former was widely perceived as bringing with it a corresponding liberation from the latter.

Punishing Atheism

Plato’s main speaker in the Laws posits a very strong link between atheism and lawlessness: ‘No one believing in the existence of the gods according to the laws has ever deliberately done an unholy deed or uttered lawless words, but only someone of whom one of the following three things is true: either, as I have said, not believing in them, or, second, thinking that they exist but do not concern themselves with mankind, or, third, thinking that they are easily appeased by being bought off with sacrifices and prayers’ (885b). Given the work’s titular theme of legislation for an imaginary city, and the further fact that that city, Magnesia, is to be constructed as a theocracy, in which the laws stand proxy for divine intellect itself, it is natural that Plato’s focus should be on the role of the gods in the legal system—both as the subjects of cults sanctioned by the laws, and as divine powers reinforcing the authority of the laws.

In addition, the full Platonic picture involves the identification of celestial gods with the soul or souls that move the heavens with perfect orderliness, and hence their proclamation as a model for humans to emulate in their own lives. In Plato’s earlier Republic the concern had been an educational one: the gods and heroes of mythology are role models for the young, so must not be portrayed as practising deceit, violence, etc. In his later work, the celestial gods have largely taken over from the traditional Olympians, and they do not behave in a recognizably ‘moral’ way, but rather constitute models of rational orderliness. Even in that guise they set a standard which, if properly internalized, makes a vital contribution to human morality. This is all under threat, in Plato’s eyes, if atheism takes hold.

Plato’s main speaker proposes precise penalties for the various forms of impiety, forms among which outright atheism is included (908b–909d). In the framing of the relevant law it is conceded that there may be atheists of naturally just temperament and behaviour. But even these, the legislator continues, are a threat to society, since they are bound to mock religious rituals, thus undermining their fellow-citizens’ faith in them. One is reminded here of an Athenian society contemporary with Plato which called itself the Bad Luck Club, or Kakodaimonistai. Scandalously, it was said to schedule its private dinners for reputedly ill-omened days, in order to make fun of the gods and the laws (Lysias, as quoted by Athenaeus 12.76.15–29). Plato seems to think that even less vocal atheists, however good their character in other respects, will have the same insidious effect on civic discipline. Their punishment is to be isolation in the city’s phrontistērion or ‘sound-mind centre’ (one of its three established prisons) for a minimum of five years, during which only members of the city’s governing council will be permitted contact with them, for the purpose of remedial instruction. After that initial sentence, they will have an opportunity to prove their return to soundness of mind; but in the absence of such proof, they will face the death penalty.

A worse kind of outright atheist is one who combines disbelief in the divine with bad moral character. Typically such people are driven by strong passions, and use cunning, including even the cynical manipulation of others by religious fraud, to gain and exploit power, perhaps even to become tyrants. Their punishment is to be total isolation in a different prison far out in the countryside, where none but slaves will have contact with them. This is apparently to be a life sentence, and after death their bodies must be unceremoniously dumped, without burial, beyond the civic boundaries.

The Cynics

So much for Plato. In the remainder of the fourth century bce there is little sign of atheism, positive or negative, beyond a handful of anecdotes in which philosophers belonging to or influenced by the Cynic movement are reported to have expressed suitably provocative doubts about recognizing and/or cultivating the gods (see Drozdek 2007: 207–14). One of these (reported by Tertullian, Ad nationes 2.2), contains a declaration of agnosticism: Diogenes the Cynic remarks that he does not know whether the gods exist, only that it is expedient that they should. But by and large, Cynicism was more concerned to challenge religious convention than to take a stance on existential questions about divinity.

Epicurus: a Crypto-atheist?

We thus move into the Hellenistic age, officially 323–31 bce, the period sandwiched between the collapse of Alexander’s empire and the start of the Roman empire. At the very end of the fourth century the first new Hellenistic school emerged, bringing atheistic issues very much to the fore. This was the Epicurean school, with its revised and updated atomism. Like his atomist predecessor Democritus, Epicurus had a tightly restricted ontology: nothing exists but atoms, void and their supervenient properties. Consequently he could not admit gods to his universe unless they were themselves composed of atoms. The evidence is very clear that in the Epicurean universe gods do exist, and that they are indeed made of atoms. However, when it is asked what this mode of atomic existence amounts to, interpreters divide into two broad parties, the realists and the idealists, with the latter interpretation in effect making Epicurus an atheist. It is to this question that we must now turn.

This heavily coded statement combines the firm assertion that there are gods with an instruction to us to conceive those gods in a way which will ‘protect’ them. Later Epicureans seem to have no doubt that their school’s founder was referring, in realist mode, to biologically immortal beings. But the language chosen at least licenses an idealist alternative, that our gods are a projection of our own thought, whose invulnerability it falls to us to ensure.

According to our best Epicurean sources, the gods are anthropomorphic beings, known to us through ‘images’ (Greek eidōla) which enter our minds directly, especially during sleep. Images are films of atoms, usually travelling at high speed from solid bodies to our sense organs and thus enabling sense-perception, but others are spontaneously formed in mid air. Some of them are so fine in texture that they make themselves known directly to the mind as the stuff of dreaming and imagining, by-passing the sense organs. It was by means of these images that, as the Epicurean Lucretius explains (5.1169–82), early mankind dreamt of superhuman-seeming beings, and attributed to them the divine properties of blessedness and eternal existence. They thus formed the generic conception of god.

Their critics regularly accused them of crypto-atheism: their ‘gods’, it was alleged, were nothing more than these streams of images. Whether the charge was accurate depends on the extremely difficult question, where the images were thought to come from. On the realist view (e.g., Mansfeld 1993; Konstan 2011), the gods themselves exist as biologically immortal beings, probably living in the spaces between worlds called intermundia in Latin, rather than in worlds like ours, which are themselves definitely held to be perishable. These gods are either made of constantly renewed streams of images, or have permanent atomically constituted bodies so fine as not to be subject to damage by any coarser bodies that may happen to pass through them. It is from their distant abodes that their images travel to our world and impinge on our dreams.

According to the idealist interpreters (e.g. Long and Sedley 1987; Obbink 1996; Sedley 2011), by contrast, such indestructible beings could not actually exist in an Epicurean universe, where all compounds are emphatically said to be subject to eventual dissolution; nor could apparently living dream figures, elsewhere dismissed by Epicurean physics as a source of illusion (Lucretius 4.722–822), possibly provide telepathic knowledge of the gods as far-distant biologically immortal beings. Rather, Epicurus’ real meaning is taken to be that we intuitively construct the gods as projections of our own moral ideal, visualizing them especially in our dreams, in which we ourselves compose them out of the constantly available images, just as we do all other dream pictures (Lucretius 4.962–1036). The gods do exist, then, but as our intuitive thought-objects. They play a vital role in ethics by representing our moral ideal. The underlying conception is of entirely tranquil beings, altogether unaffected by the fear of death which so plagues human lives, and uninterested in building and governing worlds, since such activities would detract from their sublime peace of mind, much as political involvement detracts from human happiness.

The gods are, in short, idealizations of the Epicurean way of life. They are anthropomorphic because, as idealized versions of our own human lives, they could be nothing else. They can therefore be identified with the anthropomorphic gods of popular religion, with the qualification that these latter have typically been misconceived as having additional characteristics, such as warlikeness or vindictiveness, that detract from their blissful existence. Even here, according to the idealist reading, those characteristics are projections of human thought, albeit this time morally misguided thought that wrongly attaches positive value to the wielding of power. True religion—understood as the correct conception of the ideal human existence—is to be arrived at by means of moral education.

Even according to the alternative, realist interpretation, Epicurus sides with atheism to the extent that he denies all divine intervention in the running of the world, thus claiming to liberate his followers from the fear of divine wrath. But on the idealist interpretation his position is one that in most theological contexts would be called fully atheistic, and indeed was so called by Epicurus’ own critics. Why, if so, would he not declare his atheism openly? Part of the answer may be that Epicurean communities, wherever they sprang up, relied on toleration from the local authorities, and a reputation for atheism, with its implied rejection of civic cults, would have hampered that objective. But in any case, Epicurus on moral grounds sincerely recommended participation in religious cults as a proper expression of respect for ideal beings, a stance which would have sat very oddly with an outright assertion that these beings do not actually exist.

Theodorus and Euhemerus

Two other early Hellenistic intellectuals—the Cyrenaic philosopher Theodorus, nicknamed ‘Theodorus the atheist’, and a maverick chronicler-cum-novelist named Euhemerus—came in time to be added to the standard list of atheists. The reports of Theodorus’ atheism include little detail, although intriguingly he was credited with the authorship of a book entitled On gods which was alleged to have influenced Epicurus (Diogenes Laertius 2.97). Euhemerus, by contrast, came to be a celebrated figure who exercised a subtle but pervasive influence on Hellenistic and Roman thought. ‘Euhemerism’ has in fact come to be a regular term for the policy of rationalizing myth to the point where it no longer contains any divine element. Euhemerus’ own contribution seems to have consisted in describing, apparently in the form of historical reportage, the human origin of the traditional gods Ouranos, Cronos, and Zeus on the island of Crete. This has much in common with the apotheosis theory attributed to Prodicus (see above), and it did in time earn Euhemerus the label ‘atheist’, although it is no easier than it was with Prodicus to be sure the epithet was accurate. It is important to bear in mind that the deification of human dynasts had by his day become a common political practice, and was not likely to be seen as far-fetched.

Carneades

The final figure in our story is one whose contribution to atheism has been largely overlooked. Carneades was the head of the Academy in the mid second century bce. This prestigious school, founded by Plato in the fourth century, had since the 260s bce adhered to a fundamentally sceptical stance, subjecting all philosophical positions to systematic doubt. Carneades himself wrote nothing, but his arguments were recorded by his follower cl*tomachus, and after his death they became the basis of a schism among his successors.

How did this schism come about? cl*tomachus himself understood Carneades’ arguments as serving the goal of sceptical epochē, ‘suspension of judgement’, which we may interpret as the policy of keeping all philosophical debates unresolved. This result was achieved, not necessarily by arguing against every position, but sometimes by actually defending one position in order to provide a counterweight to another. Thus for instance in order to induce suspension of judgement about determinism, which had been ably defended by the Stoics, Carneades sought to strengthen the Epicurean anti-determinist stance (Cicero, On fate 23). In cl*tomachus’ view such defences were strategic, and never represented Carneades’ own philosophical preference.

However, another close associate of Carneades, Metrodorus of Stratonicea, claimed to have been privy to the master’s own sincerely held beliefs, and rejected the epochē interpretation of his strategy in favour of a more doctrinal alternative. Metrodorus’ view influenced Philo of Larissa, who became head of the Academy in 110 and soon adopted a fallibilist policy, retrojected onto Carneades, according to which it is proper, after examining both sides of a debate, to endorse the more ‘likely’ or ‘convincing’ doctrine, so long as you acknowledge that you could be wrong.

The relevance of this background is as follows. Carneades produced a whole battery of arguments against the existence of gods, on which Cicero (a major source on Carneades, and himself a pupil of Philo) makes his Academic spokesman Cotta comment as follows: ‘Carneades used to say these things not in order to eliminate the gods (for what could be less fitting for a philosopher than that?), but to convict the Stoics of failing to settle any matter concerning the gods’ (On the nature of the gods 3.44). This interpretation amounts to a kind of fallibilism, implying that Stoic theism could well be fundamentally true but is not philosophically proven. As a historical thesis, Cicero’s remark reads like an exercise in apologetics, intended to make Carneades acceptable to a Roman audience, as is in fact confirmed by a striking parallel. Back in 155 bce Carneades had shocked Roman audiences when he made a speech in favour of justice, but another the next day against justice. Cicero (as reported by Lactantius, Divine institutes 5.14.3–5 and Epitome 50.8) adopted a similar apologetic stance with regard to this episode too, maintaining that Carneades’ aim on that notorious occasion had been not to undermine justice, merely to show that the defences of justice by Plato and Aristotle were fallible. In both cases Cicero’s apologetic justification looks like an unhistorical concession to Roman conservatism.

Cicero’s claim that Carneades’ theological arguments were nothing more than attacks on the Stoics has been generally accepted, but there are excellent reasons for disbelieving it. The later Sceptic Sextus Empiricus (Against the professors 9.138–90) has preserved for us, apparently more or less verbatim, a large number of these arguments, drawn from cl*tomachus’ full catalogue of them. Nearly all of them conclude: ‘Therefore there are no gods’. Many arguments from the same collection are presented by Cicero too (On the nature of the gods 3.29–52), but in the guise of a response to Stoic theology, with the specifically atheistic conclusions suppressed. On close examination, none of them really serves that anti-Stoic agenda. True, some of them use premises of Stoic origin, but no more numerous than those they borrow from Plato, Aristotle, poets, and other philosophers, or for that matter from common religious belief. To all appearances the arguments are aimed against a broad coalition of theists, without discrimination among them, and draw their premises from a similarly broad range of respectable sources. In response to this coalition, they aim specifically to make a case for the non-existence of the gods.

Here are some examples. The most famous of Carneades’ arguments are sorites or ‘little-by-little’ arguments, well targeted at both philosophical and popular theologies, which had alike failed to determine where the boundary of the ‘divine’ is located (see Burnyeat 1982). Between the clearly divine and the clearly mortal there lay various grades of beings, such as daimons and heroes. Likewise the divinization of major cosmic components, such as the sea’s identification with Poseidon, left the status of smaller bodies of water in doubt: rivers were generally divinized too, but in that case why exclude streams, and even seasonal torrents? Carneades’ sorites arguments thus typically took the form ‘If A (e.g., Zeus) is a god, B is a god; if B is a god, C is a god;…if F is a god, G (e.g., a torrent) is a god; but G is not a god; therefore A is not a god; but if there were gods, A would be a god; therefore there are no gods’.

Various other of his arguments start from the premiss that gods, if they exist, are ‘living beings’ (zōia). This classification was in fact common to nearly all the major philosophers, including Plato, Aristotle, the Stoics, and the Epicureans. (The Greek word means ‘animal’, but calling god an animal sounds misleadingly weird in English.) What Carneades sought to show is that (a) divine living beings would have to be endowed with at least as many senses as we have; (b) the sensory process would in its very nature involve changes and strains; and (c) these would in turn undermine god’s essential inalterability and indestructibility. Then there are dilemmatic arguments that turn on the unacceptable consequences of god’s being either finite or infinite, either bodily or non-bodily, either with or without speech, and so on. Finally, we may mention the long series of arguments on god’s moral qualities: being ex hypothesi good, he must possess all the virtues; but these will have to include, for example, courage, which implies that he is not immune to danger.

The fact that his arguments were all aimed at establishing the non-existence of the gods does not make Carneades a positive atheist. There is every probability that his anti-theistic arguments were, in accordance with cl*tomachus’ interpretation of his strategy, meant to restore a balance between the two opposed cases, that for and that against the gods’ existence. The Stoics had themselves provided a long series of syllogisms arguing the case for theism, whereas the philosophical tradition had bequeathed little on the atheist side. It fell to Carneades to rectify that imbalance. Hence from the arguments themselves nothing can be inferred about Carneades’ own theological stance.

Nevertheless, two important consequences do follow. First, what we possess, stemming from Carneades, is a very substantial catalogue of dialectical arguments for positive atheism. In the history of atheism they deserve to be acknowledged as such. Second, if we go with cl*tomachus’ historically credible interpretation, Carneades was a negative atheist as the term is understood here (that is, he did not positively affirm the existence of any god or gods), and moreover a kind of agnostic. The term ‘agnostic’ is appropriate because his ideal of suspending judgement about everything included in its scope suspending judgement as to the gods’ existence. His approach differed from conventional agnosticism in that it did not privilege theism over any other doctrinal stance as a target of philosophical doubt. Nevertheless, unusually motivated though it was, it did add up to a variety of agnosticism.

Conclusion

In the four centuries down to the beginning of the Roman empire in 31 bce, disbelief in the existence of gods was a recognizable if rare stance. Yet we know of virtually no public intellectual during that period who displayed it with complete openness. The authors of the positive atheist texts criticized in Plato’s Laws apparently succeeded in remaining anonymous. Epicurus, if, as some hold, he had a leaning toward atheism, presented it as a kind of theism. Protagoras and Carneades, for all their courage in the face of religious teachings, opted for versions of negative atheism. Theodorus ‘the atheist’ may conceivably have been an exception to this pattern, and have actually flaunted his atheism, but our evidence for him is too thin to provide confirmation.

Atheism has often been a politically, culturally, and even legally risky thesis to embrace. As we have now seen, its early history bears eloquent witness to that fact.

Bibliography