The origins of mechanics – both the practical development of devices and the creation of a body of theory about their operation – are obscure. Doubtless elements of the discipline, such as the use of lever and balance, axle-and-wheel, clepsydra and siphon, predated the fourth century bce. However, the creation of a discipline requires more than the use of certain kinds of technology: there needs to be some kind of unifying idea, some perception of commonality. Looking at the disparate elements that came to be included in the Greek discipline of mechanics, it is difficult to identify exactly what that perceived commonality was. It may be that the reason why mechanics came to be recognized as a discipline – a field of knowledge – was no more than a recognition that certain devices make possible results that would not have occurred without them, and that some of them worked in similar ways. The contribution of mechanics to the history of philosophy arose from its ensuing theories, as well as a more general commitment to the idea that the principles at work in mechanics can also be found in the natural world.

This chapter will examine the evidence concerning the creation of a discipline called ‘mechanics’ during the fourth century. Such a reconstructive project requires great caution against reading back assumptions about what ‘mechanics’ means, or about what is self-evident. In the end, I find the evidence from the fourth century to be inconclusive.

Texts have survived from the Hellenistic period concerning a so-called pneumatikē technē. I shall refer to the field as ‘pneumatics’, although – as scholars have noted – the English transliteration does not exactly capture the sense of the subject matter, which concerns devices worked by flowing water, compressed air or steam. As Landels aptly suggests, the German title Druckwerke best captures the sense: although there are some anomalies, the field is largely about what might be called pressure effects.

The characterization of pneumatics as delimited by pressure-driven devices is, admittedly, a simplification of a vaguely defined field, and it works better for Hero's collection than for Philo's. The latter – especially in the more extensive Arabic version – includes a number of water wheels, which do not depend on pressure differentials so much as the simple weight of water. Both Pappus and Philoponus describe a branch of mechanics as being concerned with devices that raise water, perhaps by analogy parallel to the category of devices used to lift weights. However, the vast majority of devices in the pneumatica collections use pressure differentials to regulate the flow of water in different ways.

I have previously discussed the classical philosophical theories offered to account for ‘rarefaction effects’. Some pneumatic devices of the Hellenistic era work by the reverse procedure, forcing air into a smaller place than it otherwise occupies. These ‘compression effects’, as I call them, are credited to Ctesibius.