work, Ma‘asei ha-Shem.6 In a philosophically informed discussion, Ashkenazi explains that their bodies “are from the hylic element that receives the forms … [enabling them to] appear at times in animal form, and at other times in the form of people. Moreover, what people have said of them—that they roam in flight—should not be cause for bewilderment, since their bodies are from a fine, simple substance.” Idolaters of old used to animate their idols with impure spirits so that they might speak. Their punishment was to become like the very impure spirits they used to animate these idols. Ashkenazi adduces Psalm 135, verse 18, “they who make them become like them,” in support of this thesis. Ashkenazi then continues to extend his structural analogy to the workings of wind instruments. Understanding their principle of amplification enables one to understand why the wicked dead demons enter a human to commandeer his or her organs of speech.
[B]odies that are from that same fine element are so highly refined that “their voices are unheard” (Psalms 19,4). Just as we see that when a sound comes to the hollow of instruments such as trumpets, rams’ horns, or the like, it will amplify and sound. So too the voice from that body, due to its delicacy, is inaudible. But when it comes to the hollow of the throat of that fallen person [ha-nofel], the throat is an apt vessel to make the voice audible, notwithstanding its fineness. Thus the body will make its voice and its speech heard in that throat, without moving the lips at all. And people commonly say7 of the fallen that an evil spirit [ruaḥ ra‘ah] has entered them, for they too call them a wind/spirit [ruaḥ].
Ashkenazi largely conflates demons (shedim) and spirits (ruḥot) in his attempt to provide a cogent “natural” explanation for the phenomenon of spirit possession.8 His analysis attempts to conform to contemporary natural philosophical notions of the nature of occult bodies, while referring repeatedly to biblical texts and contemporary dybbuk accounts to demonstrate the correctness of his thesis. His concerns were exclusively theoretical and metaphysical, and the consequences of his terminological conflation on the practice of exorcism were of no concern to him.
Yet even before such conflation, confusion reigned. The terms ruḥot, mazzikim, and shedim were used interchangeably in Jewish literature from the rabbinic era to the early modern period. This terminological fluidity was noticed by commentators such as R. Moshe bar Naḥman (“Ramban,” 1194–1270), according to whom “the shedim [were] called mazzikin in the language of our rabbis, and se‘irim in the language of scripture.”9 This imprecision notwithstanding, occasional attempts were made to impose order on these classifications. Thus the most eminent of medieval Jewish commentators, R. Solomon Yiẓḥaki (“Rashi,” 1040–1105), distinguished in his talmudic commentary between “shedim—they have human shape and eat and drink like people; ruḥin—without body or form; and lilin—human form but they have wings.”10 No term was taken to refer to a human ghost, though all partook of familiar human characteristics.
Although the identification of a possessing spirit as a ghost seems to have been common among late medieval and early modern Christians, clerical authorities generally suppressed such notions as soon as they were summoned to perform an exorcism, though they did not eliminate belief in possession by the dead entirely.11 In so doing, they were following theological traditions going back to Augustine that denied this possibility.12 Islam similarly disallowed for the possibility of possession by souls of the deceased. As a recent orthodox scholar of Islamic demonology has written, “Since the human soul enters the barzakh upon the death of the body and the state of the barzakh prevents any contact with this world, it would not be possible for disembodied human spirits to possess living beings or to communicate with them.”13 Although Christian clerics attempted to suppress the notion that the dead could possess the bodies of the living and Islamic theorists dismissed the very possibility, Jewish religious authorities came to regard spirit possession as typically resulting from just such an etiology. An exceptional position could still be found among Jewish scholars: the sixteenth-century Italian physician-rabbi Abraham Yagel rejected altogether the notion that the possessing agents were disembodied souls and argued that they were in fact demons impersonating humans. Yagel’s view was thus congruent with the one held by the religious authority of his broader cultural environment, the Catholic Church.14
Yagel’s atypical view should not obscure the fact that early modern rabbis generally believed that ghosts as well as demons could possess the living. Sometimes they were thought to do so together. Whatever the case, the exorcist had to know whom he was up against in order to work effectively. R. Ḥayyim Vital (1542–1620), the most prominent disciple of R. Isaac Luria and a major figure in what follows,15 described how one was to distinguish between possession by a ghost and by a demon:
Demons [Shedim] and a ghost [ruaḥ ra’ah] are two distinct types. For the ghost is the spirit of a person, which after his death enters the body of a living person, as is known. First one must recognize their signs: the demon compels the person, and he moves spasmodically with his arms and legs, and emits white saliva from his mouth like horse-froth. With a ghost, he feels pain and distress in his heart to the point of collapse. However, the primary clarification is when he speaks, for then he will tell what he is, particularly if he speaks after having been compelled by you with adjurations and decrees.16
In Sefer ha-Goralot (Book of Lots), a work attributed to Vital, a bibliomantic divination procedure promises to discern whether the spirit afflicting someone is a ghost or a demon.17 In this source, the demon may be a Jew, a Muslim, or a Christian. This should be no great surprise, given the traditions noted above regarding the various human qualities and appetites of these creatures, sexual among them. It is also strikingly similar to Islamic demonological views according to which demons are either Muslims or kuffars (disbelievers).18 Nevertheless, techniques designed to remove a demon were regarded as ineffective for removal of a ghost. In his unpublished work on practical kabbalah, Vital therefore provided a technique to exorcise such spirits, which, he wrote, would not work against a disembodied soul.19 Although it therefore remained possible to view demons as potential possessing agents in early modern Judaism, most sources at our disposal clearly indicate that the primary agent in nearly every case was taken to be a ghost. Even where a demonic spirit was present, its role was as guardian and tormentor of the disembodied soul. Ghosts had become the clear stars of the show.
The doctrine of gilgul provided the ideational basis for this development and lent it cultural coherence. Rather than viewing gilgul in conceptual terms alone, and in isolation from other factors, in what follows I sketch a genealogy of dybbuk possession that traverses the historical terrain vertically as well as laterally. The vertical view emphasizes gilgul and kabbalistic practice; the lateral, cognate idioms and speculative contextualizations. The resulting image is one that narrows the phenomenological gap between dybbuk possession and spirit possession in early modern non-Jewish settings without effacing their distinctiveness.
Magico-mystical techniques employed by Spanish Kabbalists to bind disembodied spirits of the dead to the souls of the practitioners were decisive in first translating doctrine into practice; indeed, no a priori reason exists to assume that the techniques did not precede the doctrine. Although such techniques may have been indebted to ancient (and contemporary) magical necromancy, their inscription in kabbalistic sources affiliates them most directly to the concept of ‘ibbur (a masculine noun, literally meaning “conception”; v. le-hit‘abber, to become pregnant; n. ‘ubar, fetus).20 Before considering such techniques, however, we must briefly survey the development of this notion21.
Although it first appeared in Jewish mystical sources as a general term for reincarnation, synonymous with gilgul, ‘ibbur appears to have been imbued with a distinct meaning of its own for the first time in Ramban’s commentary on Ecclesiastes.22 Little is known of Ramban’s views on reincarnation; the Geronese and Castilian schools of kabbalah treated gilgul as
