In this detailed and wide-ranging book, W. V. Harris brings together several different but complementary aspects of ancient thinking about dreams in a volume that is thought-provoking in the best sense of the term.
Having laid out the most important modern, scientific, theories about dreaming in the introduction, Harris' first chapter focuses on the phenomenon of the 'epiphany dream'. The advantage of Harris' use of this terminology for the dream type more commonly known as the 'message dream', is that it allows him to focus on the appearance of a divine or dead figure within the dream, rather than the form or content of the message. In this chapter, Harris offers a history of the epiphany dream from antiquity to the Renaissance, in which he connects the decline of the epiphany dream with an increase in 'secular' thinking about dreams.
The book's most interesting chapter is the second. In this chapter, 'Greek and Roman Dreams that Were Really Dreamt', Harris examines a selection of ancient dreams that are particularly prone to being understood as genuine accounts of dreams as far as the dreamer can remember them (as opposed to dreams recorded at a greater length of time and in less detail; Harris cites Calpurnia's dream before the murder of Caesar as an example of an obvious invention). He chooses to focus on St Perpetua, Artemidorus, the emperor Constantine, St Augustine and Aelius Aristides. Harris' methods for determining whether the recorded dreams were really dreamt include examining the convenience factor (Constantine's dream is deemed too convenient to be an accurate report, p116; Augustine and Monnica must have adapted a dream to suit their purposes, pp117-118); the personality of the author (Artemidorus is considered too 'gullible' to be a reliable witness, p114) and the nature of the report (Perpetua's dreams are too long and too obviously worked into a narrative p112; Aristides is the only dreamer whose dreams pass the test, as being sufficiently bizarre and dream-like, p120).
There are some obvious problems with any attempt to recover 'real' dreams from literary dream reports. In the first place, as observed above, many dream reports may be entirely spurious or fictitious. Secondly, it is difficult to remember a dream even a few hours after one has dreamt it, never mind trying to remember it weeks, months or years later. Harris offers a reasonably convincing defence of the attempt, observing that, in certain cases, the question is unavoidable if we wish to understand the text (p94) and he notes that, while he is undoubtedly dealing with probabilities rather than certainties, this is true of all ancient historical endeavours (p93). There are also clear weaknesses in some of the arguments – as Harris concedes, some scientists would argue that not all dreams are necessarily bizarre (pp18-19) - and while I tend to agree that Constantine's dream was remarkably convenient, it is not impossible that a general, immediately before a battle, might dream of a divine sign because he might be expecting to. However, as Harris notes, there are some cases (such as those of Aristides or Perpetua) in which the text cannot be approached without considering this question. Harris makes a very good case in favour of this area of research and his conclusions are probably as near as we can get to determining which of the ancient dream reports may reflect real dreams. Harris' work in this area will surely provide the basis for much future research.
Chapter Three occupies the greatest part of the book and takes us into an area which Harris has already addressed in some detail in his 2003 article 'Roman Opinions about the Truthfulness of Dreams'. Here, he expands this to include Greek and Roman opinions about the truthfulness of dreams. This section will provide an essential reference for anyone looking into this area. Although I might disagree with the extent of change in favour of belief in divine dreams that Harris sees in the second century AD, the chapter overall provides an extensive and important overview of an essential topic.
Chapter Four offers a similarly thorough discussion of 'naturalistic' (chiefly philosophical and medical) theories concerning dreams, focussing on those theories which discussed dreams as a human phenomenon. This is a most welcome contribution to scholarship on dreams, which can sometimes display a tendency to over-emphasise writers such as Artemidorus or Aelius Aristides, who were particularly interested in dreams as divine signs. In this way, scholars sometimes downplay the significant theories on dreams as a human phenomenon seen in the works of Aristotle and the Hippocratic corpus. Here, Harris is able to go some way towards redressing the balance.
Overall, this is a wonderfully detailed and fascinating work of scholarship. Perhaps one could be more guarded on occasion – it is too easy to refer to a 'credulous tradition' that survived scepticism (p152), or to talk of Aristides' 'hypochondria' (p92) (a suggestion for which there is no real evidence; there are plenty of chronic medical conditions not fully understood by modern doctors, and there would have been many more in the ancient world). But Harris is not unaware of this problem, and in particular cites the work of Keith Hopkins, who was instrumental in encouraging modern historians to empathise with the beliefs of those they are studying (p21; Hopkins 1993: 26). However, Harris' focus on scepticism and 'naturalistic' thinking is most welcome, as scholarship on dreams has too often been preoccupied with the idea that everyone in antiquity thought of dreams in the same way as Artemidorus. This is a timely study and comprises essential reading for any historian of ancient dreams.