Heard melodies are sweet,
but those unheard are sweeter
John Keats, Ode to a Greek Urn

1Musical hallucinations are an uncanny and poorly understood part of human experience, but not an especially unusual one (Sacks 2008; Sacks & Blom 2011). Many people report having had them and the related but separate phenomenon of “earworms” is extremely common (Williamson et al. 2012). Recent scientific and medical advances in the subject have often been linked to a humane clinical approach, but the clear associations between musical hallucinations and a wide range of health issues, such as schizophrenia, deafness and a number of physiological and neurological conditions, means that they are generally perceived in a fundamentally medical paradigm (Evers & Ellger 2004; Stewart et al. 2006; Evers 2010). However, while the neurological basis of such experiences is now quite well established, the meanings ascribed to them and the kind of music heard remain in some ways culturally and individually determined (Griffiths 2000). There are people who do not understand auditory hallucinations to be a malign or pathological part of their lives, as the Hearing Voices movement, led by Marius Romme and his colleagues, has been arguing since the 1980s, drawing on phenomenology and other fields (Romme & Escher 1994; McCarthy-Jones 2012). Less attention has been paid to people expressing this perspective about hallucinations with musical content, although Oliver Sacks and others have written on the subject (Sacks 2008: 5–6, 80–82; Blom & Sommer 2011: vi). Nevertheless, the fact that musical hallucinations are not necessarily experienced as a pathology is an important starting point for understanding the phenomenon.

2The pathological perspective has dominated historical understandings of musical hallucinations, reflecting deep assumptions about music, the body and health. Seeking rational, physiological explanations for such experiences has been seen as a hallmark of modernity, and the nineteenth century in particular is often considered the beginning of our own medical view of the subject and the end of older “superstitious” and religious perspectives. The historiography of musical hallucinations has hitherto understandably focused rather Whiggishly on the origins of the definitions and theoretical frameworks that have formed the basis of medical and scientific investigations of the subject for the last 200 years. The medical concept of hallucination, the distinction from illusion and the search for physiological causation that all developed up to the time of Esquirol’s work in the 1830s are still prevalent today – a primary trajectory in the important historical work of German Berrios (1990; 1996: 35-70) or Toshihiko Hamanaka (2003). This has perhaps led to an exaggerated view of the nineteenth century as a break in continuity in the discussion of musical hallucinations, but, as this paper seeks to show, the period offers a rather more complex and ambiguous picture.

The medicalization of musical hallucinations

5The term “hallucination” entered the language in the early seventeenth century through the likes of the Swiss physician Felix Platter, who used the phrase Mentis hallucinatio to mean an error of the mind. As far as hearing is concerned, he listed tinnitus aurium (tinkling), sibilus (hissing), pulsatio (pulsation) and fluctatio (fluctuation) under the heading of depraved hearing, which is when the mind “perceives the external sound falsely, and that by reason of an internal sound praeternaturally raised in the ears, outnoising this external sound, or mixing itself with it, and confounding it, and that either continually, or only by fits, besides which impediment it breeds no small trouble to a man and makes him unquiet” (Platter 1602–1608: 239–240; see Hamanaka 2003: 25). The term “hallucinations” gained wider currency after it was included in the nosology of the eighteenth-century French physician François Boissier de Sauvages, who defined them as “errors of mind originating from impairment of an extra-cerebral organ and producing deceptions of imagination”, and, again, linked them to perceptual problems caused by physiological problems like vertigo and tinnitus (Hamanaka 2003: 27–28).

7As far as specifically musical hallucinations are concerned, there is evidence that musical hallucinations were interpreted as medical conditions as far back as Galen. In his 1621 medical and philosophical compendium The Anatomy of Melancholy, Robert Burton gave an account of the musical hallucinations of Theophilus, a patient of Galen’s, who “thought he heard musick, from vapours which made his ears sound, etc.”, and that others “are deceived by echoes, some by roaring of waters, or concaves and reverberation of aire in the ground, hollow places and walls” (Burton 1838 [1621]: 283). During the Enlightenment, there were occasional references to similar phenomena in medical works, often with the newer language of nerves rather than vapours. For instance, some physicians associated musical hallucinations with what we would now call earworms and the strange workings of musical memory. The French physician Joseph-Louis Roger’s 1758 Tentamen de vi soni et musices in corpus humanum argued that “Certain diseases of the brain and nerves remind us of ideas which we believe to be forgotten, and awaken in us, on the occasion of these ideas, ancient sensations, which at the time had only little affected us”, noting the case of a “young man, much addicted to study, being sick, said one morning as I entered his room that he had not slept in two days because he had not been able to get an aria from [Rousseau’s opera] Le Devin du Village out of his head” (Roger 1803: 341). This account was still being cited on work on musical hallucinations a century later (Hagen 1837: 67).

8By the mid nineteenth century, the Esquirol view was becoming standard in medical approaches to musical hallucinations. A student of Esquirol, Jules Baillarger, produced what is often described as the first clinical study on the subject in the 1840s. It explicitly set out to explain the “physiological” causes and had a clear medical focus on the status of hallucinations as an “accompanying symptom” of “febrile delirium, delirium tremens, mania, ecstasy, hysteria, etc” (Baillarger 1846: 274). However, Baillarger was just one of several physicians in this period to discuss the topic in these terms. Another example can be found in the British society physician Henry Holland’s essay on the role of hallucination, dreams and delusion in cases of insanity in his Medical Notes and Reflections, from several years before Baillarger’s publication. With clear resonance with twenty-first-century patient testimony, it described:

A lady advanced in years, and much devoted to music throughout life, [who] became subject, without obvious cause, to impressions of musical sounds or airs, so unceasing, and often of such intensity, as greatly to fatigue and distress her, and even to a great extent to interfere with nightly rest. Sometimes the sounds were harsher and more irregular, yet still tending to assume musical time. (Holland 1840: 234)

10The medical paradigm on musical hallucinations also linked the debate on the subject to broader anxieties about the relationship between music and sickness. Such hallucinations could be associated with dangerous illness and even death, in the context of a remarkably widespread debate on the pathological consequences of musical nervous overstimulation for the body and mind (Kennaway 2010; 2012a; 2012b). The leading French physician Alexandre Brière de Boismont’s book On Hallucinations of 1845, for example, gave an account of an often-cited case:

In the month of October, 1833, a woman, aged twenty-eight, a native of Piedmont, went to the ball given at the fête of her village. She danced for three days with a kind of frenzy; after that she perpetually heard the sound of the music which had so delighted her. Fire-balloons had been let off, and of these she saw a continual succession, one making way for another. This hallucination disturbed the vital powers, and ultimately brought on a kind of nervous consumption. Dr Brosserio observed that the musical sounds continued to increase with the progress of the disease, and the woman died without ever ceasing to hear them. (Brière de Boismont 1855: 316)

Diagnosing mystical musical hallucinations

12Religious ideas had long dominated pre-modern conceptions of musical hallucinations. It is striking that God rarely appears visually in the Bible, but reveals himself in sound. For instance, when Moses sees the Burning Bush, it is described not as God but as an angel, and it is the voice of God himself who says “I am that I am”. Such auditory experiences are not limited to the spoken voice. In his Varieties of Religious Experience William James suggested that there was a strong musical element even in much verbal religious “hallucination”. “In mystical literature, such self-contradictory phrases as ‘dazzling obscurity’, ‘whispering silence’, ‘teeming deserts’, are continually met with”, he wrote. “They prove that not conceptual speech, but music rather, is the element though which we are best spoken to by mystical truth. Many mystical scriptures are indeed little more than musical compositions” (W. James 2008 [1902]: 308). Especially from the seventeenth century, older traditions of conceiving of experiences of musical hallucinations in terms of divine or demonic intervention often clashed with the increasingly dominant pathological view of the phenomenon. Many observers turned to medicine to debunk the claims of visionaries (Heyd 1995: 196–203). For instance, Robert Burton, whose account of musical hallucinations we have already looked at, blamed “visions” on “bad diet, sickness, melancholy” (Burton 1838 [1621]: 681).

14This medical approach to musical hallucinations was also applied to the past, with great interest shown in “retrospective diagnosis”. In much the same way as Jean-Martin Charcot and Paul Richer would later famously reinterpret the physical gestures associated with ecstatic religious experience as symptomatic of hysteria in their 1887 Les Démoniaques dans l’art, physicians earlier in the century were already plundering religious history for examples of musical hallucinations in order to subject them to psychiatric interpretation (Bertrand 1826; Lélut 1836; Charcot & Richer 1887). For example, the German psychiatrist Friedrich Wilhelm Hagen wrote about the musical aspects of the visions of the seventeenth-century German mystic Jakob Böhme: “When he was dying the famous Jakob Böhme heard the harmony of the spheres and called out, ‘What do I hear? What music is that?’ and asked his son if he did not hear it” (Hagen 1837: 186; see also Clemens 1858: 28). This theme of hearing divine music at the point of one’s death, or of that of a loved one, was widespread. Examples connected to John Bunyan and others come up again and again, and the association between death and heavenly music was, not surprisingly, also a common theme in later discussions of Spiritualism (Bunyan 1873: esp. 637; Brittan & Richmond 1853: 271). Elsewhere in the medical literature, such as in the essay on the theory of religious madness by the German psychiatrist Karl Wilhelm Ideler (1848), one finds similar stories about John Calvin hearing drums and a variety of forms of religious musical hallucinations associated with the likes of Heinrich Suso and the English hermit Richard Rolle (Godwin 1987: 72–73).

15However, not all physicians took this reductive and exclusively negative view. In an ostensibly similar vein, Brière de Boismont described the musical hallucinations of the German mystic Hans Engelbrecht: “[H]e heard, during a period of forty-one nights, angels sing and play celestial music; and could not help joining with them. The persons with him were so transported with joy that they sang with him during a whole night.” De Boismont put it into the context of psychopathology:

The state of ecstasy being a phenomenon of extreme nervous excitability, it is clear that it must be manifested at all periods when the mind has been agitated by fanaticism, and by a belief which brings in its train either ardent hopes or strong fears. Hence it is evident that it has been more prevalent in times of ignorance, than at those periods in which a more advanced civilization has enabled reason to triumph over imagination. This remark is undeniably true; and a very little erudition is required to acknowledge the existence of ecstasy in the pythonesses of antiquity, in those initiated into different mysteries, the famous sects of the Middle Ages, the possessed, the convulsed, Shakers, Illuminati, et cetera. (Brière de Boismont 1855: 215)

16However, Brière de Boismont, as a good Catholic, was keen to preserve a space for genuine religious revelation, writing that “it is essential to establish a distinction of the highest importance between the ecstasy that I shall call physiological and a morbid ecstasy” (Brière de Boismont 1855: 216; 1839; T. James 1995: 147–50). In contrast to many of his contemporaries such as Louis-Francisque Lélut and Alfred Maury, he insisted that hallucinations by themselves were not proof of madness (Lélut 1856; Maury 1845). Medicine had a role to play in explaining the physiological version, but he also emphasized the possibility of benign hallucinations, partly guided by willpower. Thus, even in mainstream medicine, the continuities with ambivalent early modern views of musical hallucinations between medicine and mysticism were clear, and the nineteenth century looks less like a caesura in the development of thinking on the subject.

The persistence of non-pathological interpretations in non-establishment discourses

19Especially in Germany, the debates within Mesmerist circles took on an increasingly occult edge as the nineteenth century wore on, and a major theme in their discussion of musical hallucinations relates to “rapport” or telepathy. The descriptions of musical hallucinations in this context incorporate them into broader metaphysical models, eschewing pathology, in much the same way as those in a more straightforwardly religious context. The positive tone that was generally adopted on the subject in Mesmerist circles was due in part to their determination to show that animal magnetism was not just a freakish sideshow, but a useful way of looking at the world. The 1836 book Theorie des Somnambulismus oder des thierischen Magnetismus (Theory of Animal Magnetism) by the Hegelian philosopher and theologian Johann Ulrich Wirth was one of many to give examples of magnetized people having musical hallucinations via telepathy (Wirth 1836: 180). And Heinrich Schwarzschild’s Magnetismus, Somnambulismus, Clairvoyance (Magnetism, Somnambulism, Clairvoyance) from 1854 talked of the harmony of the spheres heard by clairvoyants while they mystically perceived the solar system (Schwarzschild 1854, Vol. 2: 75). Nor were such experiences a matter only for fringe periodicals and eccentrics. In 1818 the very mainstream Blackwood’s Edinburgh Magazine related a similar story about a boy called Mateus Schurr, who was inexplicably able to hear and recall music played at a great distance (anon. 1817–1818: 441–442).

20Another discourse that was by no means as marginal as one might think, and which also made up a substantial contribution to discussions of musical hallucinations, was phrenology. Partly because phrenologists were keen to demonstrate their ability to use their knowledge to stimulate musical hallucinations via the skull, they were also remarkably positive about the experience. The usual phrenological explanation of musical hallucination was an “over-excitement of the organ of Tune”, that is to say, an excess of stimulation in a particular part of the brain, the relative size of which was supposedly detectable through the skull. For instance, in the Edinburgh Phrenological Journal and Miscellany in 1829, an account of what might be seen as Auditory Charles Bonnet Syndrome noted that “Several cases are on record of persons deaf and dumb perceiving and being delighted with music … The organs of Tune and Time are largely developed in Mr Burns [the subject of their research]” (A.C. 1828–1829). No suggestion was made that these symptoms were pathological.

21Similarly, the Lyon physician Alexandre Bottex’s 1836 Essai sur les hallucinations discussed musical hallucinations as a boon to be explained phrenologically: “[I]ndividual organs of animal life become active, and so necessarily the feelings and thoughts dependent on these organs awake too; only in this case can these actions occur without the influence of the will. If only one organ is active, the dream will be simple. One embraces the object of his love, another hears beautiful music” (Bottex 1836: 24) This phrenological approach, while in some ways discredited in its own terms, would set the scene for later work on the localization of musical experience and hallucinations. The decline of phrenology and Mesmerism over the nineteenth century in the face of scientific questioning had the effect of reinforcing the clinical approach to musical hallucinations, but other forms of more non-pathological and positive conceptions of the experience thrived in the context of Romanticism.

Pathology and creativity: Musical hallucinations among artists

22Bottex’s suggestion that musical hallucinations could be a positive experience is linked to the widespread sense that they were linked to musical creativity among composers. This chimed with very ancient conceptions of creativity as the result of literal external divine inspiration akin to the Muses, while also fitting with the developing medical discourse of mad genius. In the debates on the pathological character of hallucinations in France in the 1840s, Brière de Boismont, who attacked the crude pathologization of the phenomenon, himself cited the “internal hallucinations” of musicians as a key part of their art (Brière de Boismont 1856: 297). Later, the leading psychiatrist Richard von Krafft-Ebing suggested that there was an overlap between musical imagination and hallucination. He wrote that “Composers may base the intricacies of their instrumentation and timbre of their compositions on particularly acute and lively reproductive powers of their acoustic memory. That individuals with such extraordinary sensual gifts are more likely to hallucinate than those with poorer imaginations who function more in the realm of abstract ideas can be admitted without further ado” (von Krafft-Ebing 1890: 117). Others, such as Friedrich Wilhelm Hagen, were at pains to distinguish hallucination, which he saw as essentially pathological, from healthy “pseudo-hallucinations” (Hagen 1868; Hirsch 1896: 90).

23These hallucinations or pseudo-hallucinations were often taken seriously and became part of the discourse of Romantic genius. The idea of divine inspiration via hearing voices had generally been more prominent than musical hallucinations in mystical traditions, but for nineteenth-century artists the musical version seemed to offer more associations with creativity, in a tradition going back to Socrates’ daemon. While hearing voices was becoming more and more limited to psychiatric discussions in the nineteenth century, “hearing music” still carried a touch of glamour, even if it too had associations with sickness. Artists, composers and musicians, and perhaps especially their acolytes and biographers, often found the idea of divine inspiration via musical hallucination of one kind or another irresistible. For instance, the English Spiritualist Georgina Weldon was a friend of the composer Charles Gounod, and she described his composition process as essentially hallucinatory:

I am a good musician, and yet I cannot conceive what he experiences when he composes. He seemed to hear something with his eyes closed, over which he would keep his hands clasped for a few minutes. My husband and I would then keep strict silence whenever we saw him thus inspired, for fear of rousing him from his dream. At the end of that time he would say, “It is beautiful! I have got it all!” He had heard, in that short space of time, hours of music, which he immediately wrote down, as if something had copied (on some organ which none of us possess) photographed sounds, and which he declared he heard played by an orchestra composed of such beautiful sounds, that nothing on earth could give the faintest idea of their exquisite sweetness and power. (Weldon 1882: 129)

24The nineteenth-century fascination with musical hallucinations in terms of the uncanny process of musical creativity perhaps also gives us a clue as to why after 1800 there was such a boom in both medical and literary interest in the subject. Along with the idea of music overstimulating the nerves, this link to inspiration became almost a cliché in Romanticism. Conversely, Romanticism’s conception of music in terms of extreme feeling, sometimes even amounting to madness, was in marked contrast to the orderly aesthetics of the Enlightenment and surely provided a key aspect of the basis of the idea of musical hallucinations as both a heroic inspiration and a pathological menace. In Romantic opera, it comes up time and time again, notably in Bellini’s La Somnambula and Wagner’s Tristan und Isolde. Heinrich von Kleist recorded his own memories of a full orchestral hallucination on the Rhine in a letter of 1800, and made it a theme in his story Saint Caecilia or the Power of Music. Likewise, in Victor Hugo’s Les Misérables, the character of Cosette “thought she heard a noise. It was a hallucination produced by the melancholy and magnificent chorus of Weber, which lays open before the mind terrified depths” (Hugo 1887: 99).

26In the context of composition, the most famous example of this kind of glamorous pathological musical hallucination is naturally Robert Schumann, who was the subject of a pathography by Paul Julius Möbius (Möbius 1906). The true character of his mental illness has been fiercely contested – bipolar disorder, schizophrenia, syphilitic General Paralysis of the Insane, et cetera, et cetera, have all been put forward over the years (Ostwald 1985). In any case, in the early 1850s Schumann began to hallucinate a single note, then music that he described as “glorious, and with instruments sounding more wonderful than one ever hears on earth”. Later he heard specific pieces, such as Ein’ feste Burg ist unser Gott by Bach. As in many accounts by psychiatric patients, the hallucinations started as pleasant to a certain extent. He talked about it being a source of “exquisite suffering”, but it soon took on a sinister aspect that left little scope for denying its pathological character, even if it also retained a questionable Romantic appeal (Ostwald 1985: 4).

27Another fascinating example of the interaction between medical and Romantic conceptions of musical hallucinations is provided by Moreau de Tours’ discussion of the topic in his book Du Hachisch et de l’aliénation mentale (1845). Although he was interested in genius as a pathological condition, his descriptions of drug-fuelled musical hallucinations are remarkably positive. The sense of hearing, he argues, is rendered extraordinarily impressionable by the action of hashish. To give a sense of its impact, Moreau quoted at length the writer Théophile Gautier, a member of the Club des hachichins, which also counted Baudelaire, Dumas, Balzac and Nerval among its attendees, and to which Moreau famously acted as drug supplier. Gautier’s quotation describes a scene in which someone is playing music from Weber’s Der Freischütz on the piano. The combination of hashish and music leads to a pleasant complex synaesthesia with musical and visual hallucinations.

“My ears”, said he, “had prodigiously developed; I heard the sound of colours; Green, red, blue, and yellow sounds came to me by perfectly dishonest waves. An overturned glass, a cracking arm-chair, a low pronounced word, vibrated and resounded in me like rolling thunder; My own voice seemed so strong that I dared not speak of fear of overthrowing the walls or burst like a bomb; More than five hundred pendulums sang to me the hour of their flute, copper, and Argentinian voices. Each object touched made a note of harmonica or harp wind. I swam in an ocean of sonority floating, like islands of light, some motifs of Lucia [di Lammermoor] or the Barbier [de Séville]. Never was such a beatitude hinted at me with its efflorescence.” (Moreau de Tours 1845: 71–72, quoting Théophile Gautier; see also T. James 1995: 98–129)

28The ambivalence one seems to find on musical hallucinations in many different contexts is therefore evident in debate on their role in creativity. While very different from Brière de Boismont’s claims for non-pathological creative hallucinations, Schumann’s case offered an archetypal model of the suffering hallucinating genius, destroyed by the same thing that created his talent, which nevertheless provided a (highly paradoxical) positive angle on the subject. Similarly, Moreau de Tours also offered scope for positive interpretations of musical hallucinations beyond the pathologizing paradigm, with artistic inspiration in the place of mysticism.

*

29It is thus clear that rather than marking an entirely new era in which musical hallucinations were increasingly limited to discourses of medicine and pathology, the nineteenth century saw a remarkably fluid, ambivalent and contested view of the phenomenon. This can be seen among physicians themselves, from Brière de Boismont’s defence of non-pathological religious revelation to Moreau de Tours’ positive depiction of Gautier’s experience on hashish. The mystical interpretation may have faded among established churches, but it flourished among widespread non-establishment religions and influential groups with often disputed scientific credentials. While the achievements of the scientific study of musical hallucinations must be acknowledged, an examination of the real breadth of the debate on the subject in this period shows a much more varied and upbeat view.

30A broader cultural approach to nineteenth-century thinking on musical hallucinations also reveals obscured commonalities. In particular, it shows that the new Romantic ideology of music, with its interest in excessive emotional states, the sublime and madness, helped lay the foundations for both a cult of creative musical hallucinations and the idea that they were pathological. Romanticism’s often quasi-mystical conception of inspiration made it the heir to religious traditions with more positive interpretations of musical hallucinations, but it was also influenced by medical views of the role of pathology in creativity. In turn, it is clear that Romantic ideas also influenced medical views of the phenomenon, especially in the genre of pathography, which combined an interest in the role of morbidity in musical talent with an inability to resist the glamour of degenerate genius.

31Finally, an approach to the history of musical hallucinations should try to match the way that the Hearing Voices project and the thoughtful interpretations of Oliver Sacks have involved listening to patients’ views in order to gain a more nuanced grasp of the experience. Rather than focusing only on those who suffered from such hallucinations or accompanying symptoms, the broader view of musical hallucinations set out here can more easily include those who showed no discomfort or even enjoyed their “symptom”. For them, both in the past and today, the pathologizing paradigm does not apply, as it would not have applied to the “hypochondriac” described in a 1750 book by the Scarborough physician Peter Shaw, “who imagined he constantly heard Seraphic Music; and instead of gratifying the Physician for his Cure, accused him, as a Thief, who had robbed him of his Delight” (Shaw 1750: 114; see Forrer 1960; Sacks 2008: 85). A view of musical hallucinations that is based only on the pathological angle will not get to grips with the full range of such experiences or their history.