Distinguishing between dreams and visions in ninth-century hagiography*

by Margaret Kenny


Today the dream and the vision have acquired definitions that distinguish them as wholly separate phenomena. The dream is an internalized experience that occurs during sleep,1 while the vision is a waking event that is apparently perceived otherwise than by ordinary sight.2 The question ‘Can we distinguish between the Byzantine dream and vision’ then arises from a need to understand their role in society. Byzantine oneirology suggests dreams and visions are hierarchically structured, each having a defined function and, from this, a role in society. A review of ninth-century hagiography, however, reveals that the dream and the vision cannot always be so clearly distinguished.


The study of dreams falls into two areas, the study of dream interpretation and the study of dreams. I wish to obscure, a little, the demarcation lines between the two, for I wish to access the content of the dream, without entering into a dialogue that interprets the dream imagery, as a means of assessing its role. The initial research was completed using the Dumbarton Oaks Hagiography Database for the ninth century. It is designed to allow a systematic search of a large number of Vitae for specific topics, and I have located some 322 episodes from 58 vitae. I will look closely at two of the episodes.


We cannot divorce ourselves from our own culture when examining another, and our exploration of Byzantine material will necessarily be conditioned by our own cultural world. That we should find their dream and vision material perplexing is unsurprising for within the last century there have been very many theories regarding our own dreams, most of which disagree radically. Recently the trend of dream study is one of fusion;3 of creating an inter-theoretical discipline that enables the dream to be studied from a number of perspectives. The theories that have influenced this paper are Psychoanalysis,4 Analytical Psychology,5 Content Analysis,6 Neuroscience,7 Lucid Dreaming8 and Cognitive Psychology.9 On the surface they may appear incompatible bed-fellows but they all seek a similar goal, which apart from dream interpretation, is to understand the function of dreams and to ascribe them a value. Such a fusion then facilitates an inclusive classification of six dream-types. They are:

  1. Personal-mnemic – which includes everyday matters in the dreamer’s life
  2. Medical-somatic – which includes those episodes related to the workings of the body
  3. Prophetic – which present aspects of future events
  4. Archetypal-spiritual – in which the dreamer explores existential questions, and which results in some transformation of behaviour
  5. Nightmares – with upsetting or frightening images
  6. Lucid dreams – in which the dreamer is aware of experiencing a dream and then consciously alter the events.

Dreams are pan-cultural, however: while dreams may be universal to humans,10 it does not follow that each culture will assign the same value to each dream-type. Thus while the personal-mnemic dream carries the greater influence within a ‘western’11 culture, this should not be expected of other cultures, present or historic.

Byzantine oneirology

All cultures appear to experience the same dream-types although their emphasis of status is different. The Byzantine interest in classifying the dream-vision experience is evidenced through the continuous evolution of their oneirology.12 While it evolved over the centuries, it maintained the division of phenomena between the body/soul experience of the dream that occurred in sleep, and the soul/divine communication of the vision that could occur in either a state of sleep or wakefulness.13 The dream experience was viewed as ambiguous, and yet, while considered highly suspicious, dream interpretation, or oneiromancy, proved perennially popular.14

Byzantine oneiromancy

The function of oneiromantic literature was to provide an interpretation of the dream. This was achieved through the analysis of the dream imagery in conjunction with the social status, economic activity and gender of the dreamer. The resulting interpretations centre entirely on the secular and physical elements of the individual – that is their health, wealth and status. Any figures that populated these dreams, known or fictitious, or alive or dead, sent messages that were related to the daily activities of the dreamer. For example in the Oneirokritikon of Nikephoros ‘If you dream of seeing a naked woman you will see your wife’s grave’.15 Or in the Oneirokritikon of Achmet ‘if someone dreams that a priest entered his house and fell asleep on his bed, he will become friends with that man’.16 Oneirokritika are orientated toward the personal-mnemic, and occasionally, medical dream-types, and any elements of prophecy originate within a body/soul context. Thus, as universal interpretations they would not be applied to archetypal-spiritual episodes. References to dream interpreters and dream-books are frequent, but not as being used by the holy or the saint. There is never a suggestion that they were used in a religious context.

The material

To turn to the episodes. One is taken from the life of Niketas of Medikion and one from the life of Theophano. Both of these episodes involve a visitation. Neither the visitors nor the visited are saints, but in each case a saint was the reason for the episode. Thus while only indirectly involved, these holy figures necessarily have an impact on the context and function of the experience. Involving the saint elevates the episode from the level of everyday considerations, to that which explores existential concerns such as ‘Who am I’ and ‘Where do I fit in relation to society and the holy’. Thus they can be classified as archetypal-spiritual episodes for they transform the material life of the recipient as a direct result of the experience.

The episode from the life of Niketas is generally identified as a dream, whilst the episode from the life of Theophano is identified as a vision. The basis of this, and all such identification, is the vocabulary used in these episodes.17 Modern definitions would have oneiros denote a dream that occurs in sleep18 and hupar as a waking vision.19 However, I believe that things are not as clear cut as they may first appear, and these two episodes have been chosen because they highlight some of the difficulties involved when making such a distinction.

Niketas of Medikion20

In 813 Niketas of Medikion became the Hegoumenos of a Bithynian monastery. During the persecutions by Leo V he was exiled but was later recalled under Michael II. He died in 824. The dreamer is a fellow monk, Nikolaos, who blaspheming against Niketas then receives a dream.

‘In a dream his own father (who died before this) stood near him and threatened him. ‘Withdraw’ he said ‘from the slaves of God’, thenceforth the unruly one behaved more sensibly, he no longer annoyed the righteous one and even prevented others from doing this.’


Theophano was the first wife of Leo VI. Chosen by brideshow and married at fifteen, she was also imprisoned with Leo. She died aged about thirty and after her death performed miracles for the family of her Hagiographer.

In the life of Theophano the author tells us that at one hour past noon, at the onset of sleep, ‘In a vision Martin the Artoklines approaches him’ with a request that the author write hymns praising her (Theophano’s) miracles, when the author professes his disinclination to do so, a long conversation between the two ensues. ‘How do you say you love me, but refuse these small requests?’ And to him I responded ‘What are these, O revered friend?’ And he said; ‘I said to praise the saint in different hymns and as far as God will make it possible for you to tell, to weave a canon to the glory of the Lord and to her honour and remembrance’. I replied to him ‘What basis of her virtues shall I use? She possesses no ascetic pathway, no struggles of martyrdom, display of miracles. I am at a loss what to say or write’. But he said to me ‘Because of God do not shrink away from this undertaking, but taking paper and pen write what I tell you.’ Taking pen and at the same time paper, joyfully to him I said; ‘Say what you wish; I am ready to write as much as you address to me.’ He said to me; ‘Do it as far as ‘Christ set you in place for the church as a luminous lantern which in times past served as a guide”. And I awoke again, bearing in my memory the word which had been spoken.’


Specific vocabulary was available to the author from which to create a particular environment for an episode. For the vision an author could use hupar, optasia, apokalupsis, chresmos, fainoo and ekstasis, and of the 322 episodes under investigation these appear in 52. Should the author wish to emphasise the appearance of a messenger then fasma and epifainoo were available, and these can be found in 54 of the episodes. For a dream environment, oneiros establishes this context, and of the 322 episodes this is used in 72. If the author declines to be so specific, syntax is utilized to create a more nebulous context that could be either dream or vision. hupnos and koimaoo tell of sleep while enupnios tells of an appearance in sleep. An alternative, much favoured by authors, was to suggest sleep and then incorporate words related to sight, such as horoo22 and even blepoo.23 Very often to mix the context further, the author can refine this technique by suggesting a dream environment but incorporating specific vision vocabulary. For example a dream in which an angel appears, as happens in the vita of David, Symeon and George.24

Therefore, even with the available vocabulary the distinction between the dream and vision is not always entirely clear. This is illustrated by the narratives from the vita of Niketas and of Theophano. Each account creates a similar scene in which the recipient is visited by a familiar figure who persuades him to undertake a specific course of action. For Nikolaos this involves the protection of Niketas from his adversaries, while our reluctant Hymnographer is persuaded to take up his pen and write. This poses the question: should these two be separately classified?

The building blocks of dreams

Apart from the vocabulary does the context create a distinction between these phenomena? There are generally seven components, which in any number of combinations will comprise the framework of a narrative account.25 These are time, place, the mental state of the recipient, the dream/vision figure, the message/scene, the initial reaction to the episode and the action taken as a response to it. These components provide the building blocks of the episode, but only rarely supply evidence of its origin, and in our two episodes not at all.

The time and place play only a minor role in setting the scene, but if the remaining elements are ambiguous and the vocabulary nebulous, this can play a deciding factor in determining what the episode is. In the dream (oneirooi) of Nikolaos, we are given no time or place of the episode. As a dream, this would imply a night-time experience but the audience is left to infer this. As for Theophano’s reluctant Hymnographer, he tells us he receives his vision at the onset of sleep, one hour after midday. However, the vocabulary used is hupar and this is generally understood to denote a waking-vision. Here the time and place only add more smoke to an already cloudy issue.

The mental state of the recipient often holds a key to the length and the detail of the episode, but alone it is insufficient to determine the origin of it. Thus, while the audience is left to ponder the depth of relationship between Nikolaos and his father, it is the appearance of the father that suggests an origin of the experience. The same can also be said of Theophano’s disinclined author. It is with the messenger in both of these episodes that the demarcation lines begin to blur. We are told that Nikolaos has a dream and that the Hymnographer has a vision, and yet they both receive a visitation. The Hymnographer is able to identify his visitor as Martin, uncle of Theophano and friend to himself. The audience remain uninformed as whether he is alive or dead, but the author makes the context clear. This is a vision. Nikolaos is also able to identify his visitor, for it is his father. Moreover, the audience are informed that he was dead when the communication took place. As such, his journey must have commenced in either heaven or hell. In light of the message – as a means of enabling the protection of the blessed Niketas – this implies a heavenly origin. As for the messages themselves, each has a similar purpose. In both, the visitor commands a course of action be adopted which ultimately the recipient carries out.


How then are these two episodes, and many more like them, to be distinguished? Or can it be that they were not expected to be so rigidly segregated? We must take seriously these narrative accounts and their ability to be readily understood by their audience, both in terms of content and function. Nikolaos and the Hymnographer undergo a similar experience and therefore this implies the accounts were expected to receive a similar reception.

These episodes function on several levels, but for each the role is of validation. On one level they serve to validate a transformation of behaviour by the recipient. For Nikolaos this entails he not only desist in his slanderous activities, but that he actively engage in protecting Niketas from others who continued to attack him. For the reluctant Hymnographer the transformation moves him from disbelieving inactivity to enthusiastic authorship about Theophano. On another level they validate the saintliness of Niketas and Theophano, while on a third level they validate the cult that was to grow around them.

My review of ninth-century hagiography does not confirm and maintain the theoretical structure of Byzantine dream-vision classifications. Rather, syntax and content are used in such a way as to blur these distinctions, creating a tension between theoretical hypothesis and practical application. Authors use dream vocabulary within contexts that are clearly vision narratives. These episode are elevated above the value of the everyday dream. Its content and context removes it from the level of the personal-mnemic dream to that which Jung would have called a ‘big dream’26 and that McFague would call a ‘root-metaphor dream’,27 making them a religious experience or event. In doing so the distinction between the dream and vision becomes so blurred as to be lost altogether. The ultimate result of this technique is that the personal internalized experience, that is, the dream, is displaced. The vision has not merely become the more important, it has become the prevailing experience in its relation to the saint.


* This paper was given as a communication at the thirtieth Spring Symposium ‘Byzantium dead or alive’ at Birmingham in March 1996.

1 C. Rycroft, The innocence of dreams (London, 1979); also J. Empson, Sleep and dreaming (London, 1989).

2 The new shorter english dictionary, ed. L. Brown (Oxford, 1993), 3589.

3 K. Bulkeley, The wilderness of dreams (New York, 1994).

4 S. Freud, The interpretation of dreams, 1900 tr. J. Strachey (New York, 1965), and also Introductory lectures on Psychoanalysis 1917a tr. J. Strachey (New York, 1966).

5 C. G. Jung, The Archetype and the collective unconscious 1951, tr. R. F. C. Hull (Princeton, 1969), and also ‘General aspects of dreams’ 1948a in Dreams, 2nd ed. tr. R.C. F. Hull (Princeton, 1966).

6 C. Hall and R. Van de Castle, The content analysis of dreams (New York, 1966), also C. Hall and V. Nordby, The individual and his dreams (New York, 1972).

7 J. Hobson, The dreaming brain: how the brain creates the sense and nonsense of dreams (New York, 1988), see also the work of Dement and Rycroft et al.

8 S. Laberge, Lucid dreaming (Los Angeles, 1985).

9 H. Hunt, The multiplicity of dreams: memory, imagination and consciousness (New Haven, 1989).

10 W. Dement, ‘The effect of dream deprivation’, Science, 131 (1960), 1705.

11 This is a difficult term and is representative of an ethos allied to economic development rather than a geographical linguistic boundary. For example it includes Western Europe and North America.

12 Beginning with the first Christian oneirology by Tertullian and ending with the last Byzantine oneirological text by Manuel II Palaiologos.

13 Synesios of Cyrene, John Klimakos, Anastasios of Sinai and later Michael Psellos, all accept this distinction.

14 G. Calofonos, Byzantine oneiromancy, Mphil thesis (Birmingham, 1995), charts the progress and popularity of oneiromancy as well as its sources of opposition, both religious and legal. S. Oberhelman, The oneirocritic literature of the late Roman and Byzantine eras of Greece, PhD thesis (Minnesota, 1981), provides a translation of the seven surviving Oneirokritika, tracing their inter-dependance and the manuscript tradition.

15 Oneirokritika of Nikephoros, 120, tr. S. Oberhelman, The oneirocritic literature of the late Roman and Byzantine eras of Greece, PhD thesis (Minnesota, 1981), 143.

16 Oneirokritika of Achmet, 149, tr. S. Oberhelman, The oneirocritic literature of the late Roman and Byzantine eras of Greece, PhD thesis (Minnesota, 1981).

17 This is the basis of distinction used in the Dumbarton Oaks Hagiography Database for the ninth century and also in Greek-English lexionaries.

18 Liddell and Scott, Intermediate Greek-English Lexicon, 7th ed (Oxford, 1992), 559.

19 Liddell and Scott, Intermediate Greek-English Lexicon, 813.

20 Acta Sanctorum, April 1, 3rd edition, ed. J. Carnende (Paris and Rome, 1866), XXVI, 38, 17-18; also Dumbarton Oaks Hagiography Database, record card 21362.

21 E. Kurtz, Zwei griechische Texte über die HL. Theopano, die Gemahlin Kaisers Leo VI. (St. Petersburg, 1989), 21-22, chapter 30; also Dumbarton Oaks Hagiography Database, record card 15945.

22 144 episodes in the database create a vision context by using words for sight, of seeing something – horoo.

23 There are 6 episodes in the database that use blepoo: Key 15892, 16582, 18643, 12874, 8129, 15728.

24 David, Symeon and George – Dumbarton Oaks Hagiography Database Key 20442: an angel appears in a dream (onar).

25 J.S. Hanson, ‘Dreams and visions in the Graeco-Roman world and early Christianity’, Aufstieg und Niedergang der römischen Welt, 11:23, 1396-1427, provides an examination of early Christian material, and while episodes in Byzantium suggest a similar construction, a definitive study has yet to be undertaken.

26 Jung, The Archetype and the collective unconscious 1951, tr. R. F. C. Hull (Princeton, 1969).

27 S. McFague, Metaphorical theology: models of God in religious language (Philadelphia, 1982).