Chapter 1
Visual hallucinations: history and context of current research

G.E. Berrios1 and Ivana S. Marková2

1Department of Psychiatry, University of Cambridge, Cambridge, UK

2Centre for Health and Population Sciences, Hull York Medical School, University of Hull, Hull, UK

1.1 Introduction

'Visual hallucination' is the name of a class of heterogeneous phenomena which share as a common feature the report that a subject (the 'hallucinator') is entertaining a putative 'visual' image of something placed in the public space that others cannot see. Hence 'absence of an object to be seen' has become part of its conventional definition. 'Images' may range from simple dots to complex forms in movement. This chapter will deal only with the latter. The hallucinator's report can be interpreted as meaning that he is entertaining a visual image or that he only believes that he is so doing, as it might well be the case with psychotic or obsessional hallucinations (Castilla del Pino, 1984; Gimenez, 2000; Fuentenebro and Berrios, 2000).

Disambiguating such reports is not always straightforward. On occasions, it can be reasonably claimed that proxy representations of the image in question can be ascertained by technical means (e.g. neuroimaging) as may well be the case in regard to so-called 'organic' or 'provoked' visual hallucinations (Berrios, 1985; Manford and Andermann, 1998; ffytche, 2007). However, when the proxy representations are counter-intuitive and merely correlational, that is, when changes do not seem to be related to those brain regions obviously associated with visual perception (as it seems to be the case with some visual obsessional or psychotic hallucinations) (Boksa, 2009) then disambiguation becomes problematic. The possibility that the hallucinator is only expressing the belief that he is entertaining a visual image cannot be easily dismissed (Castilla del Pino, 1984; Gimenez, 2000).

Therefore, it is likely that the class 'visual hallucinations' is but a mixture of clinical phenomena which are different from each other, both from the phenomenological and aetiological point of view. Indeed, this was the way they were perceived until the 'unitary' view was first proposed during the early nineteenth century (Berrios, 1996). Historical accuracy requires that two 'unitary views' are distinguished: UV1 refers to the view that ALL hallucinations, regardless of their sense modality, are structurally the same phenomenon; and UV2 refers to the view that all visual hallucinations, regardless of the disease in which they appear, are the same phenomenon. Ever since UV1 and UV2 became the official view, voices have been raised against them (for a detailed analysis of this literature see: Ey, 1957, 1973; Lanteri-Laura, 1991).

Before the nineteenth century there was no unitarian view of visual hallucinations (UV2). Visions, apparitions, phantoms, fantasies, imaginings, contemplations, supernatural insights and so on were all conceptualized as different phenomena in their own right (Tyrrell, 1943; Green and McCreery, 1975; Berrios, 2007). Some of them were considered to be pregnant with meaning to represent forms of communication between god, unknown forces and man or between men themselves. By the early nineteenth century, in what became the scientific approach to hallucinations, these experiences were shelled out of all semantic content, lumped together and treated as mechanical 'disorders' of the brain. Although the semantic view of hallucinations still lingers on, both in folkloric Western psychology and in some non-Western cultures, it cannot be said to have the force of the scientific view.

The current general concept of hallucination and that of visual hallucination were both constructed during the early nineteenth century (Ey, 1939; Berrios, 1996). Although there have since been debates on their aetiology, the conventional assumption remains that hallucinations in general are a disorder of perception (Berrios and Marková, 2012), and the same applies, mutatis mutandi, to visual hallucinations. This has made debates on the nature of visual hallucination dependent upon changing theories of visual perception (Hamlyn, 1961).

It remains to be seen whether UV2 should be considered epistemologically superior to earlier views, that is, whether it is more helpful both to the understanding and management of these phenomena. This can only be decided by empirical research. Unfortunately, no one seems interested in carrying out studies comparing UV2 against earlier taxonomic and aetiological views of visual hallucinations.

By the end of the nineteenth century, the current 'operational' definition had been put together (Berrios, 2005). Visual hallucinations: (i) were to be defined as reports of real visual images of unascertainable public objects - not of the belief that the person was experiencing an image and (ii) which regardless of their phenomenology or clinical associations resulted from pathological changes in the visual system. Agencies external to the body (such as spirits or ghosts) were no longer to be considered as part of the explanatory model. 'Personality' and 'culture' were allowed within the explanatory model but only to account for the 'content' of the visual hallucination (i.e. thus according to their 'culture', hallucinators may see a devil, a dragon or whatever).

From the above, at least three conclusions can be drawn:

1. The phenomena currently called 'visual hallucinations' do not seem to be ontologically stable 'natural kinds' but will-o'-the-wisp phenomena, which in each historical period have been configured in a different way.
2. To 'understand' these phenomena properly (and hence to manage them) a set of tools is required, which may be more complex than any offered by the neurosciences and,
3. To discharge their brief, we will need to sketch the history and epistemology of: (i) visual hallucinations and (ii) the social activity called 'research' and explore their historical interactions.

Historical analysis shows that UV2 has become a hindrance to the understanding of visual hallucinations. Exploring their neuro-mechanics or mapping them in the current language of cognitive neuropsychology is bound to be insufficient. Per contra, this chapter will put forward the view that: (i) before empirical research starts in earnest much work on the clinical phenomenology and taxonomy of visual hallucinations needs to be done and (ii) a new taxonomic approach will show that 'visual hallucinations' names a rag bag of heterogeneous phenomena, each of which will require a different aetiological account. Lastly, because subjects suffering from these clinical afflictions may need direct clinical intervention, conceptual and ethical auditing become particularly important in this type of research.

1.2 The construction of visual hallucinations

1.2.1 History

The history of visual hallucinations can be broken up into four periods. The first period extends from Classical times to the beginning of the nineteenth century (Berrios, 1996). At that point, a physiological period started in the work of Johannes Müller (Berrios, 2005). A hermeneutic period developed after the 1850s and is characterized by work on dreams, parapsychology and the reappearance of the moot question of 'hallucinations in the sane' (James, 1995). The organicist period starts with the work of Tamburini (Berrios, 1990a, 1990b) and can be said to continue to this day. To some extent, these four periods run into each other (Géraud, 1989).

1.2.2 Classical period

Reports of visions and apparitions can be found very early in history (Calmet, 1641; Dufresnoy, 1752; Amat, 1885). Often considered as epiphanic occurrences, that is, as divine revelations, these phenomena were explored with tools borrowed from the theological analysis of the soul and its relationship with the deity (Bona, 1676). The resulting taxonomy, metalanguage and hermeneutic approach provided the model for the later philosophical analysis of visual hallucinations.

St. Augustine wrote: "Three kinds of visions take place; one with the eyes, when you see the actual letters; another with the human spirit, by which you think of your neighbour even though he is not there; a third with the attention of the mind, by which you understand and look at love itself . let is call the first vision 'bodily' . the second 'spiritual' . the third one, finally, 'intellectual'" (Lit. Meaning. Genesis. 6. Book XII, 6.15,16). This analysis was to be repeated many times in the history of Christian theology until its culmination in the great works of Calmet (1641) and Bona (1672).

That apparitions and other 'visual' experiences could be empirically studied was first suggested by Augustin Calmet (1641), a Benedictine monk who proposed that: (i) apparitions of angels, demons and souls were real enough, (ii) the mechanisms involved remained obscure and (iii) God had left to humans the task of finding out the said mechanisms.

Cardinal Bona (1676) in turn differentiated visions from apparitions: the former were images of subjects known, the latter of subjects unknown. There were three types of...