Demons and Douglas: Applying Grid and Group to the Demonologies of the "Testament of

Solomon"
Author(s): Sarah L. Schwarz
Source: Journal of the American Academy of Religion, Vol. 80, No. 4 (DECEMBER 2012),
pp. 909-931
Published by: Oxford University Press
Stable URL: https://www.jstor.org/stable/23357993
Accessed: 27-01-2019 20:50 UTC

JSTOR is a not-for-profit service that helps scholars, researchers, and students discover, use, and build upon a wide
range of content in a trusted digital archive. We use information technology and tools to increase productivity and
facilitate new forms of scholarship. For more information about JSTOR, please contact support@jstor.org.

Your use of the JSTOR archive indicates your acceptance of the Terms & Conditions of Use, available at
https://about.jstor.org/terms

Oxford University Press is collaborating with JSTOR to digitize, preserve and extend access
to Journal of the American Academy of Religion

This content downloaded from 132.208.246.237 on Sun, 27 Jan 2019 20:50:31 UTC
All use subject to https://about.jstor.org/terms
 Demons and Douglas: Applying
Grid and Group to the
Demonologies of the Testament
of Solomon
Sarah L. Schwarz*

The Testament of Solomon (TSol) is filled with ideas about demons,

and its textual tradition is complex and difficult. This study applies
Mary Douglas' grid/group analysis to the demonologies of the TSol,
arguing that the insights from her anthropological approach can help
us make sense of the twin challenges of this material. By situating par
ticular demonologies found within the TSol materials in Douglas' the
orized quadrants, we can make suggestions about the social locations
that might have produced them. In doing so, we can bolster both our
understanding of these demonologies within this textual history and
gain insight into the ways attention to demons might enhance our
understanding of religious communities and their social worlds over
time.

'Sarah L. Schwarz, Colorado College, Department of Religion, 14 E. Cache la Poudre St.,

Colorado Springs, CO 80903, USA. E-mail: sarah.schwarz@coloradocollege.edu. My thanks are
always due to Ross Kraemer, Robert Kraft, and John Gager for all their mentoring of my work on
Testament of Solomon. I am grateful to Susan Marks and Debra Bucher for reading and
commenting on key drafts of this particular article in progress. I also thank the participants in the
Religious World of Late Antiquity session at the 2010 SBL meeting, whose work inspired me to
revisit Douglas and demons.

Journal of the American Academy of Religion, December 2012, Vol. 80, No. 4, pp. 909-931
doi:10.1093/jaarel/lfs072
© The Author 2012. Published by Oxford University Press, on behalf of the American Academy of
Religion. All rights reserved. For permissions, please e-mail: journals.permissions@oup.com

This content downloaded from 132.208.246.237 on Sun, 27 Jan 2019 20:50:31 UTC
All use subject to https://about.jstor.org/terms
910 Journal of the American Academy of Religion

IN LATE ANTIQUITY, the world was filled with demons. From the
everyday demons who threatened average folk with fever and other
quotidian forms of harm to the cosmic forces aligned with true evil to
the philosophical understandings of demons elucidated by sophisticated
elite thinkers, there seemed to be no doubt that demons existed, only
diverse representations of their nature and their influence. Faced with
this variety, scholars of the demonic have taken many paths toward
enriching our understandings of demons in their various historical set
tings.1 This article seeks to build on their work.2 By employing the
anthropological model described by Mary Douglas as "grid/group anal
ysis," it seems that diverse demonologies might be able to tell us some
thing about the social setting(s) which produced those demon stories.
In fact, application of this model to one particularly complex set of
demonologies suggests that we might not only use demonologies to
help us conjecture about the social settings that produced them, but
that furthermore we might use those demonologies and their theorized
settings to help make a claim about textual development for a particu
larly challenging textual tradition. Through this process, we can see that
stories about demons, while inherently interesting, are also often partic
ularly revealing about communities and their deeply held values, espe
cially as they conflict with others in their social and cultural world.
There is perhaps no text more replete with demons than the so
called Testament of Solomon (TSol).3 The story, as it is represented in
Chester McCown's eclectic critical edition (1922) and Dennis Duling's

'Examples are too numerous to catalogue exhaustively, but just a few studies of demons that
have been particularly influential on the thinking behind this article include Smith (1978), Stewart
(1991), Brakke (2006), and Frankfurter (2006).
2Peter Brown's essential work must be acknowledged as crucial inspiration for this project. His
essay (1970) applied insights from Douglas' pregrid/group work to demonologies of late antiquity
and his scholarship continued to be influenced by her ideas (as in the discussions of demons in
Brown 1978). This article is not on entirely new ground, but builds on those insights and takes
them into a somewhat new direction through consideration of TSol.
3As a result of the intensely demonological focus of the text, most previous scholarship has
considered its demonologies in some fashion. The starting point for modern scholarship on the
text is Chester McCown's (1922) critical edition and introduction, which considers demons
throughout but especially on pages 43-46, noting their varied nature in the material. Duling's work
represents another key moment in the modern study of this text, as his translation (1983) and
related publications (1984), 1988) made the materials (and their demons) accessible to a wider
audience. Recent works by Sarah lies Johnston (2002), Philip Alexander (2003), and Todd Klutz
(2005) have also importantly explored the demons as part of their considerations of this material.
This study builds on their work, but instead of seeking the demonology of the original form of the
TSol, this project accepts the tradition as fundamentally composite and uses that feature as starting
point for assessing its demonologies.

This content downloaded from 132.208.246.237 on Sun, 27 Jan 2019 20:50:31 UTC
All use subject to https://about.jstor.org/terms
 Schwarz: Demons and Douglas 911

translation (1983), begins with the construction of Solomon's Temple

in Jerusalem. Solomon's favorite, a young boy associated with the
workers on the Temple, reports that the demon Ornias has been
sucking his life force out of his thumb, and moreover stealing half his
wages. King Solomon discovers this and prays to the archangel Michael
for assistance. Michael grants Solomon a ring engraved with the divine
Seal, and Solomon uses the ring to subdue Ornias and cause him to
bring Beelzeboul, the prince of demons, before him. Similarly subdued
by the power of the ring, Beelzeboul promises that he will compel all
his subordinate demons to appear bound before Solomon. This inaugu
rates a formulaic series of encounters between Solomon and the
demons, each of whom reveals his or her name, sphere of harm, and
the name of the angel who thwarts them, and then the demons are
each in turn conscripted into laboring on the construction of the
Temple. As Solomon subdues each demon, information is included for
readers—presumably "magical" practitioners—to use in their own
encounters with the demons in question. There are a number of sec
tions of the tale which seem to alter this pattern somewhat, such as
when a group of thirty-six demons appears (the decans, or degrees of
the zodiac—for more on these see below); the story of the old man and
his recalcitrant son, in which the demon Ornias reveals he can see the
future; the interaction of Solomon and the Queen of Sheba; the story of
Ephippas, the wind demon; and so on. The modern editions of the
story then conclude with Solomon's decision to sacrifice locusts on
behalf of foreign gods, at the behest of his foreign wife, and then
repenting of his idolatrous sin by writing his testament.
On one level, TSol is an extracanonical expansion of the power of
the biblical king Solomon and a folktale about the construction of the
Jerusalem Temple with the assistance of demonic labor. On another
level, it presents an encyclopedic miscellany of practical instructions for
writing amulets, curing ailments, and controlling demons. It incorpo
rates a testamentary framework—a genre known from Jewish and
Christian examples in which a great figure from the past reflects on his
sins in facing death—and draws on biblical and parabiblical literature
broadly in its details; yet it is at home with a panoply of Greco-Roman
and Egyptian notions of power, from detailed knowledge of astrology to
the power of goddesses like Hekate to specifics like the efficacious
notion of the "Ephesia Grammata" (for more on this, see below). From
dating this set of materials to assessing its status as "Jewish" or
"Christian" to evaluating its complex manuscript tradition, the
Testament of Solomon presents challenges to our categories on almost
every level.

This content downloaded from 132.208.246.237 on Sun, 27 Jan 2019 20:50:31 UTC
All use subject to https://about.jstor.org/terms
912 Journal of the American Academy of Religion

THE CHALLENGES OF "THE" TSOL

While others have argued for the relative incoherence of the

Testament of Solomon4—a complex array of manuscripts, with no singl
clear Urtext,5 in which a biblically inspired folktale-like plot is inter
mingled with specific ritual instructions—this article considers th
diversity of the late stage of the TSol manuscripts as an important da
point in and of itself. While the main story contained within this
textual tradition varies in the ways it details Solomon's use of a magic
ring to summon the demons and force them to do the work of building
the Jerusalem Temple, the miscellaneous array of ritual instructions an
other materials contained within this vary still more from each other
and from manuscript to manuscript. Any effort to answer simple que
tions about this material—such as questions of dating, provenance, etc
—requires that we first begin to grapple with the problem of what w
mean when we speak of the (or perhaps a) Testament of Solomon.
The TSol as found in modern editions and translations comes from a
range of Greek sources which vary significantly from each other. These
manuscripts are quite late, and those that contain all the various ele
ments of the composite tale (including the "demons building the
Temple" motif and testament genre as evidenced by Solomon's reflec
tions on his sins at the end of his life) date to the fifteenth to eighteenth
centuries. And yet we know that the tradition of Solomon's power goes
back at least as far as the Bible, and his power over demons could be
cited widely in antiquity. For example, Josephus tells the tale of a Jewish
exorcist making use of a Solomonic incantation,6 the Dead Sea Scrolls
contain references to Solomon's power over demons7 (as does the Nag
Hammadi corpus),8 and the New Testament also transmits the idea that

4For just two examples, McCown described the Testament of Solomon as filled with
"superstitious puerilities" (1922: 1), and von Nordheim (1980: 187) called it a "thoughtless
perversion" of the testamentary genre (translation by Duling 1988: 100).
5Contra Klutz (2005), who reconsiders the manuscript evidence and argues that manuscript
P should be considered the best exemplar of the Testament.
6Antiquities 8.2.5 (Josephus and Whiston 1987).
7In 11Q11, for example, the notion of Solomon's power over demons is invoked. However, it is
worth noting that nowhere among the preserved scrolls do we find evidence for the story of
Solomon building the Temple with demonic assistance (Garcia Martinez 1996).
8Such as On the Origin of the World 107.3, which alludes to demons listed in a "Book of
Solomon" and the Testimony of Truth 9, 70.5, which tells a story of Solomon building Jerusalem
with demonic labor and imprisoning demons in waterpots inside it—to be released by the
conquering Romans in the future. Finally, the Apocalypse of Adam 5, 78.30-79.6 uses the image of
Solomon controlling an army of demons (translations available in Robinson 1990). While this
might seem on the surface as evidence for dating the TSol to a relatively early period, there is no
reason to conclude on the basis of such references that a fully developed tradition like the TSol we

This content downloaded from 132.208.246.237 on Sun, 27 Jan 2019 20:50:31 UTC
All use subject to https://about.jstor.org/terms
 Schwarz: Demons and Douglas 913

Solomon was a source of demonic control.9 From this, we can conclude

that the traditions contained in the TSol include elements which could
be quite old—at least as early as the first century BCE—and yet the final
compilations might best be dated quite late. Furthermore, the varied state
of preservation of this tradition, in which the diverse manuscripts appear
to be so different that it is difficult to speak of a single or uniform origi
nal text,10 suggests that the people who collected and transmitted this
material cared more about efficacious information than about consis
tency or clarity. In light of the challenges presented by this material, I
have argued that the earliest elements of this tradition are the Solomonic
incantations, which might in fact have been used by ritual practitioners
(2007). Later, these elements were joined to a plot—the Temple building
tale—and perhaps even later, to a genre—the testament framework.
Making a claim for any theory of textual development is inherently
fraught with assumptions. When multiple varying manuscripts exist, any
attempt to assess how those manuscripts came into existence and how
they relate to each other is laden with the presuppositions of the inter
preter or interpreters. While seams and disjunctures in a text may appear
problematic to some (modern) readers, we generally know little or
nothing about the producers and users of late antique texts—for whom
these apparent textual problems may have been intentional and/or
expected. On the other hand, sometimes external clues or internal
markers are suggestive enough to make risking such a claim tempting in
spite of the methodological pitfalls implicit in the endeavor. Standing
behind this argument is the view that it would be extremely difficult to
imagine a single author at a single moment in time producing a text
which looks like the TSol in its late forms. While authors can adopt a
variety of techniques to lend authenticity and interest to their literary
productions, on the basis of both literary analysis and social logic it does
not appear that is the best explanation of the text which lies before us
here. Examples drawn from the TSol show cases in which there is a

find in the medieval manuscripts existed to be cited by the Nag Hammadi corpus. Instead, such
references attest to Solomon's demonological prowess, but do not enable us to conclude that
anything like the TSol in its medieval or modern forms yet existed.
See, for example, Matthew 12:22-42, which appears to be an exorcistic story hinting at
competition (between Solomon and Jesus) for the title "Son of David."
10The oldest datable physical evidence for the TSol is a papyrus fragment dated paleographically
to the fifth or sixth centuries CE (P. Vindob. G 29 436, G 35 939, G330 [Daniel 1983]). This
fragment contains elements of chapter 18, which is a section of the TSol which would stand out
even without the evidence of this copy, and in fact this free-standing papyrus might confirm that
the contents of this chapter (a so-called decans list) circulated independently of the rest of this
Solomonic material as late as the fifth or sixth century. For more on the possibly independent
character of this papyrus witness, see below.

This content downloaded from 132.208.246.237 on Sun, 27 Jan 2019 20:50:31 UTC
All use subject to https://about.jstor.org/terms
914 Journal of the American Academy of Religion

switch from first to second person, for example, or there is a break in the
flow of story, where the telling of one tale seems to be interrupted by
another story. While it is possible to imagine an author deliberately
switching from first to second person to give the illusion of an inserted
source, or self-consciously breaking up plot elements to add to the
mystery of a text, in all, this does not appear to be the most convincing
model for the creation of this document. Rather, the clues present in this
text, including radical shifts in emphasis and apparent point of view,
characters popping up and disappearing within a few lines, and various
garbled bits of what sound like ritual instructions mixed in almost ran
domly, all of which vary in their treatment among the different manu
script families, suggest the work of different editors and compilers
bringing together various independent sources over time. On the basis of
literary analysis of what we see in the medieval TSol manuscripts and of
anthropological analysis of the demonologies we find reflected therein, it
appears the TSol as it stands in its final forms would be very unlikely to
have been created by a single individual at one time.

DOUGLAS' GRID AND GROUP

As the TSol collection is replete with demonologies, ideas about

demons provide an excellent lens for seeing the varieties contained
within the late versions of this complex tale. Insights from the world of
anthropology, and particularly from Mary Douglas' work on the rela
tionship between cosmology and society, suggest that there is a correla
tion between social location and cosmological beliefs. Close attention to
the demonologies present in the TSol shows that it is impossible to
speak of one single underlying demonology—instead, source texts with
their own individual demonologies appear to have been joined together
by a compiler (or various compilers over time) whose demonology
(-ies?) was accommodating enough to enable him or her to bring them
all together. Indeed, Douglas' insights about the impact of different
social settings on the production of different ideas about demons
provide support for the idea that within the final TSol what we find are
different elements joined late in their history.11

"Although Douglas herself would not approve of this approach. Kraemer (1992) describes
Douglas' skeptical response to her project of applying grid/group theory to women's religions in
the Greco-Roman world. As Kraemer notes, "Douglas is adamant that while we can predict
cosmology from the social structure, the reverse is fraught with methodological minefields, since
different social structures can generate similar cosmologies, for different reasons" (Kraemer 1992:
20, citing Douglas 1978: 41). Clearly, like Kraemer, I disagree, and by anchoring my suggestions
about the social location of tradents of this material in data drawn from what we know o

This content downloaded from 132.208.246.237 on Sun, 27 Jan 2019 20:50:31 UTC
All use subject to https://about.jstor.org/terms
 Schwarz: Demons and Douglas 915

Douglas' key insight about the demonic and its role in society
comes across most clearly in her book Natural Symbols,12 in her eluci
dation of the categories she names grid and group. She proposes these
two key categories that govern human experience, resulting in four pos
sible outcomes. The horizontal axis, which she calls group, measures
how strongly the individual is associated with the group and how much
pressure the group can exert on individual members to conform. To
assess the degree of group strength, Douglas writes that we should:

consider how much of the individual's life is absorbed in and sustained

by group membership.... The strongest effects of group are to be found
where it incorporates a person with the rest by implicating them together
in common residence, shared work, shared resources, and recreation, and
by asserting control over marriage and kinship. (Douglas 1978:16)

Group describes how high the barriers are between the group and
the rest of society, or conversely, how freely a person can move from
within the group to outside.
The vertical axis represents what Douglas calls grid, which measures the
degree of regulation, or in contrast, the degree of autonomy, that individuals
experience, as well as the degree of implicitly common language and symbols
(which Douglas calls "classification") that they share. Thus, grid "establish[es] a
dimension on which the social environment can be rated according to how
much it classifies the individual person, leaving minimum scope for personal
choice" (Douglas 1978:16). Douglas concedes that low grid is more challenging
to assess, but "to estimate it, one could ask how freely a person disposes of his
own time, of his own goods, chooses his collaborators, chooses his own clothes
and food" (1978: 16). Furthermore, a low grid situation will allow greater indi
vidual autonomy, and make fewer categorizations based on attributes over
which people have no control (such as gender or race).
The intersection of these two axes, grid and group, creates four pos
sible quadrants into which a given community within a larger society
can be categorized. This mapping works for smaller social units, rather
than societies writ large, because as Douglas explains,

If I speak of group, then, though the group may be ever so big, so that
all the members cannot possibly know each other well, there would

Byzantine spellbooks and their users (see below) I believe we can tentatively begin to build from
the theoretical model toward the historical (if hypothetical) human actors.
"Douglas presents multiple versions of this schema in her publications, including changes
between the 1970 and 1973 editions of Natural Symbols and a somewhat different version in
Cultural Bias (1978). For a roadmap to some of the varieties, see James V. Spickard (1989).

This content downloaded from 132.208.246.237 on Sun, 27 Jan 2019 20:50:31 UTC
All use subject to https://about.jstor.org/terms
916 Journal of the American Academy of Religion

have to be in all parts of it a pressure from face-to-face situations to

draw the same boundaries and accept the alignment of insiders and
outsiders. A unit such as "England" or "the Catholic Church" would
not qualify as "group" in this sense. (Douglas 1982a, 1982b: 15)

Douglas also argues that the boundaries between the quadrants are
real, and thus "it is not possible to stay in two parts of the diagram at
once" (1982a, 1982b: 4). Furthermore, she writes that "Whatever else may
be changing, the four extreme grid/group positions on the diagram
are liable to be stable types, steadily recruiting members to their way of
life which is inevitably a way of thought. ... I claim that four distinctive
types ... are continually present, inexorably drawing individuals into their
ambit" (5). Thus, these quadrants represent ever-present ideal types of
social groups, and as Kraemer notes, "Rarely, if ever, does one of these
four combinations suffice to describe the experiences and outlook of all
members of an individual society" (17). Kraemer goes on to cite Douglas'
observation that "the social experience of most people in every society is
likely to be that of insulation, isolation, and atomized subordination
(strong grid and weak group)" (1992: 17, citing Douglas 1978: 21).
Therefore, this mapping provides a useful tool for exploring the interaction
between society and cosmology, even though it may represent an idealiza
tion which elides some of the particularities of certain examples.

SITUATING TSOL MATERIALS INTO DOUGLAS' SCHEMA

Producers and transmitters of TSol materials could have come from
three of the four categories Douglas outlines.13 It is difficult to imagine
a weak group/weak grid community drawn to the TSol, since this is the
setting in which individualism is paramount and people ascribe their
fate to personal accomplishments and failures, rather than to divine will
or a malevolent cosmos. In this situation, the very notion that good and
evil exist as abstract categories would tend to be questioned, as this
social location is characterized by relativism and a lack of absolutes.
This most closely resembles the free-market ideology of contemporary
American culture, which de-emphasizes the role of race and gender, for
example, in favor of an ideology of personal accomplishment. This
setting might be used to describe contemporary scholarly readers of the
TSol, who can approach these materials untroubled by their various
inconsistencies and find value in them for historical and academic

"Catherine Bell's chart in Ritual: Perspectives and Dimensions (1997: 45) and Kraemer's chart
(1992: 15) were both very helpful for characterizing the details of Douglas' approach.

This content downloaded from 132.208.246.237 on Sun, 27 Jan 2019 20:50:31 UTC
All use subject to https://about.jstor.org/terms
 Schwarz: Demons and Douglas 917

reasons, totally apart from their ability to provide an avenue to power

or connection to the divine. In antiquity, this was a relatively rare
classification, but perhaps could best be used to describe some
philosophers—in any case, there is little reason to identify early creators
or tradents of TSol materials with individuals in this type of group.
On the other hand, the other three options seem useful for under
standing strata within the TSol tradition. The most relevant category is
weak group/strong grid, in which people experience "insulation and iso
lation from sources of authority and power" (Kraemer 1992: 16). This is
the setting in which concern for the efficacy of ritual practices would be
the most paramount, because individuals in this situation have no
access to the standard avenues to power. Because of their alienation
from the power structure, people in this setting tend to view rulers as
arbitrary and amoral, and thus demons would tend to be viewed as

Grid and Group and Demonology

(based on the work of Mary Douglas)

Strong Grid

Weak Group, Strong Grid Strong Group, Strong Grid

• Fixed social status • Status fixed

• Weak group support • Strong group support
• Strong pressure to • Lots of rules and lots of
follow rules concern about sin as
• Arbitrary cosmos violation of rules
• No rewards, must fulfill • Good rewarded, evil
set station in life punished
• Hierarchy affirmed
Example: TSol 18, in which demons
cause individual harm and are
Example: TSol 1, in which the actions of
thwarted instrumental ly demon Ornias serve God's plan
Weak
Group Strong
Weak Group, Weak Grid Strong Group, Weak Grid Groups

• Social status can be • Status can be achieved

achieved • Sectarian situations
• Individualism • Strong group boundaries
• Egalitarianism • Lots of divisions
• Neutral cosmos between/among groups
• Personal achievements, • Dualist cosmos
not divine plan
Example: TSol 7, in which powers of the
Example: Contemporary readers of "other"
the are demonized dualistically
TSol demonologies, who are largely
unconcerned about demons

Weak Grid

This content downloaded from 132.208.246.237 on Sun, 27 Jan 2019 20:50:31 UTC
All use subject to https://about.jstor.org/terms
918 Journal of the American Academy of Religion

arbitrary and amoral, yet powerful, beings. Since the demons here are
not evil, they are rather viewed as annoyances like bad weather or crop
failure (dangerous and capable of harm, but lacking deliberate malevo
lence or the capacity of reason). Thus, abstract categories of good and
evil would be less relevant here, and instead people in this situation
would tend to be more concerned with harm from capricious forces
rather than a malevolent cosmic plan. The details of this cosmology
most closely resemble the material in the decans list preserved in TSol
18, for example. In this case, individual demons appear, each charged
with a particular sphere of harm, such as "chill(s) and shivering"14 or
"pains in the kidneys,"15 and each is thwarted by a specific ritual action
which seems to invoke the appropriate thwarting angel instrumentally.
There is no evidence in this type of material within the TSol of any
interest in the wider problems of good and evil, or with any notion of a
divine plan. Rather, the focus is on amelioration of individual problems
with practical, everyday solutions, where the impersonal universe dem
onstrates little concern for the plight of individuals.
The next category Douglas describes is strong group/weak grid. This
is the category best used to describe sectarian situations, as it encom
passes a total "us vs. them" cosmology, with extremely high barriers to
the outside world. These individuals tend to see the world as currently
engaged in a contest between the forces of good and evil, in which the
outcome is not yet decided. Groups in this classification would also
tend to possess rites to protect against invasion from outsiders, because
outsiders are viewed as the source of pollution. This is a setting akin to
Qumran, where the group members saw themselves as a beleaguered
remnant beset by outsiders, in an unjust universe where ritual was con
cerned with expelling demons or interlopers from the social body.
Some of the sections of the TSol which seem most focused on demoniz
ing the religion and rituals of the other would tend to arise in this
setting. In this scenario, elements of other traditions are seen as real,
powerful, and efficacious, but also chaotic, evil, and malevolent—so that
ultimately the religion of the nearby other is treated as a demonic force,
to be controlled through "proper" ritual.
Perhaps the most concrete examples of this are the passages con
cerning the stoichiea (appearing in 8.2 and 18.1-18.2) and the decans
(18.4-18.42), in which once-neutral astrological forces are recast as
demons subordinate to Solomon's power and control using the ring of

uTSol 18.18, MSS HLPN with minor variants.

l5TSol 18.14, MSS HP.

This content downloaded from 132.208.246.237 on Sun, 27 Jan 2019 20:50:31 UTC
All use subject to https://about.jstor.org/terms
 Schwarz: Demons and Douglas 919

divine power.16 Still another type of example emerges in TSol 7, in the

wind demon Lix Tetrax. This demon appears as a whirlwind, hurling
dust at Solomon, until Solomon spits on the ground and seals him with
the ring. Solomon then begins interrogating him about his sphere of
harm and thwarting angel, finally commanding him to hurl stones to
the top of the Temple under construction. The terms "lix" and "tetrax"
are drawn from the so-called Ephesia Grammata, a well-known and
widely used group of efficacious words going back at least as far as the
fourth century BCE (McCown 1923; Kotansky 1991: 111). "Magic"
words, or voces, were evidently considered to be powerful in their own
right, but there is no reason to think that they were typically under
stood as names of individual figures. As McCown puts it, "For the liter
ary tradition, then, the Ephesia Grammata are words of power, and
there is no evidence of their hypostatization" (McCown 1923: 132). The
places where they do appear personified are in popular, practical usage,
such as in amulets and incantations. However, in the examples found
by McCown, where they are personified in amulets, they are presented
as "beneficent, protecting spirits" (McCown 1923: 135). This is con
firmed by Karl Preisendanz, who points to the combination of Lix and
Tetrax used in connection with wind on a Cretan tablet in a benevolent
sense (1959). Therefore, the choice made by the composers of the TSol
to take two of these words of power, unite them, and make them into a
demon name, represents a particular and interesting move in the
textual tradition.
As McCown writes, "One can have little doubt, therefore, that for
the non-Christians of the fourth century A.D., as for the Cretans of
the fourth century B.C., the Ephesia grammata were still living and
active spirits, as powerful and beneficent as in the time of their pre
historic origin" (McCown 1923: 139-140). The producers and trans
mitters of TSol also viewed Lix Tetrax as powerful and active, and
thus recontexualized this spirit as a demon, subject to God's power
through Solomon and the proper thwarting angel. This pattern of
demonization, of taking elements from close traditions (such as the
surrounding Greco-Roman or Egyptian cultural context) and

l6Stoicheia, a Greek term often translated as "a fundamental component," and decans, usually
glossed as one of the thirty-six spiritual entities each connected with ten degrees of the zodiac, are
technical zodiacal terms. Since stoicheia appears in the Pauline letters (Gal. 4:3, 9; Col. 2:8, 20),
scholars have sought parallels to elucidate this term, and thus have noted the appearance of
stoicheia in TSol materials, apparently associated with decans. While the textual complexities of the
TSol make it difficult to use evidence from it to clarify matters of dating, it is noteworthy for the
purposes of this argument that these technical astrological terms have been personified as demons
within the TSol and perhaps also in the Pauline corpus.

This content downloaded from 132.208.246.237 on Sun, 27 Jan 2019 20:50:31 UTC
All use subject to https://about.jstor.org/terms
920 Journal of the American Academy of Religion

recontextualizing them as demonic (regardless of who does it), does

not tell us anything about, for example, the way the power of Lix
Tetrax was really understood by others in this period. Rather, it tells
us a great deal about the people who produced this material and
placed it in the TSol. The fact that the religion of the other is demon
ized, rather than ignored or mocked, demonstrates that at some phase
in the development of the text, this religion was real to the people
writing about it. Furthermore, the tradents wanted to use the power
of such concepts rather than discredit them, as evidenced by their
efforts to incorporate all this into a Solomonic framework. Taking the
real details of religious traditions and demonizing them is a way of
appropriating. This material is suggestive of groups in close contact,
and yet of high concerns about the integrity of group boundaries,
such that it was meaningful to demonize outside religion at the same
time as it was being appropriated.
In this category, all pollution from the outside is viewed as demonic
attack, and another facet of this cosmology might be reflected in the
way various goddesses—representative of the religion of the other, at
least at some point in the compositional history of the TSol—are incor
porated in the TSoFs purview, but are dealt with as demons. For
example, Enepsigos (who may be Hekate) in TSol 15 might represent a
dualist approach to this powerful figure, here presented as purely malev
olent and yet espousing a christological prophecy about Jesus' imminent
arrival within the passage.17 This has the effect of rending her totally
subordinate to the religious system espoused by the author or composer
of this section, and thus dealing with the inevitable demonic pollution
of the outside religion by subordinating it within the dualism of the
cosmos. Thus, in the strong group/weak grid quadrant, we find multiple
examples of highly dualist thinking, in which the presence and activity
of demons is viewed as a manifestation of the cosmic battle between
good and evil, and the cosmologies of other groups can be recast and
thus understood within a framework of demonology.
Finally, Douglas' category of strong grid/strong group describes a
setting in which there are clearly articulated social roles and a hierarchi
cal social organization with strong boundaries between inside and

''Briefly, the three-headed Enepsigos resides near the moon, and claims she can change form
and is taken for a goddess. Despite her boast, "This Temple cannot contain me" (TSol 15.6),
Solomon binds her and, while bound, she prophecies about the destruction which will come when
Jerusalem is conquered, and "the implements of the Temple which you are making shall serve
other gods" (TSol 15.8). This will result in an unpleasant period during which demons will roam
freely in the world, "until the Son of God is stretched upon the cross" (TSol 15.10).

This content downloaded from 132.208.246.237 on Sun, 27 Jan 2019 20:50:31 UTC
All use subject to https://about.jstor.org/terms
 Schwarz: Demons and Douglas 921

outside. However, because this is the social location in which priests or

other leaders are often situated, in this setting, leaders are envisioned as
living in a moral world with cosmic consequences, where good is
rewarded and evil is punished. In the view of individuals in this case,
although good and evil forces might battle, good will always win, in
accordance with God's plan. A number of elements of the final TSol
would seem to fit into this category. For example, individuals in this
location would tend to view sin as a violation of formal rules.18 This
could account for the section of the TSol (26.4-26.5) in which Solomon
falls madly in love with a foreign woman, and in order to win her
agrees to sacrifice locusts to the gods Raphan and Moloch—a sin which
leads to God withdrawing his power from Solomon, and thus the king's
decision to write a testament to warn others to avoid such transgres
sions. In this passage, the emphasis is placed on the idolatrous nature
of Solomon's sacrifice of locusts, because in a strong group/strong grid
situation, formal rules are important and moral behavior of leaders is
seen as significant. By placing emphasis on Solomon's failure to uphold
the precepts of Judaism, breaking the commandments by sacrificing to
another God, this layer of the text identifies formal rules as crucially
important. Moreover, this idolatrous sin is brought into greater relief
through contrast with Solomon's role as king, leader of the Jewish
people, and builder of God's Temple, which are again features that
would be important to individuals invested in a vision of leaders who
are morally righteous. The focus in this quadrant on clearly articulated
social roles could also be associated with the focus on the proper role
for demons in their hierarchy (or hierarchies), as demonstrated, for
example, by the attention to titles like "Prince of Demons" (associated
with Beelzeboul). These elements of emphasis on formal rules and
delineation of hierarchies appear most often in the narrative frame of
the TSol collection, in those elements which serve to advance the devel
oped plot, and thus suggest that this layer of the collection may come
from this kind of social location.
Finally, because people in this setting tend to believe that the uni
verse is fundamentally just, they might tend to emphasize the ways in
which the demons fit into the cosmic order. In this quadrant, demons
are never envisioned as working in opposition to God in dualist
fashion, and the demons will never, ultimately, win out over the forces
of good. In the case of the TSol, we see examples of this in two primary

18As Douglas writes in Natural Symbols, "Strong grid and strong group will tend to a routinized
piety towards authority and its symbols; beliefs in a punishing, moral universe, and a category of
rejects" (1982b: 63).

This content downloaded from 132.208.246.237 on Sun, 27 Jan 2019 20:50:31 UTC
All use subject to https://about.jstor.org/terms
922 Journal of the American Academy of Religion

ways: those passages in which demons are described as instruments of

divine justice, advancing God's plan even through bad behavior or
exacting penalties on God's behalf; and those passages in which
demons are described as under divine control, ultimately susceptible to
God's power. For example, the demon Ornias appears to work to
advance the divine plan in narrative framework of the TSol. In the
opening of the tale (chapter 1 of the TSol), Ornias is malevolently
sucking the life force from a young boy, but this action serves to put
the demon in contact with Solomon, which leads Solomon to gain the
power of the ring and the ability to command demonic labor in con
structing the Temple. Thus, even the demon's negative action serves a
larger positive purpose. Similarly, another consequence of the world
view held by individuals in this quadrant might be the idea that
demons and the harm they cause are ultimately mitigated because it is
all part of God's plan, so that even demonic labor can be harnessed for
a positive end such as building the Temple.
Using Douglas' model also enables us to make some very tenuous
suggestions about the social location(s) of the people who might have
generated and collected the materials within the final TSol. Her theory
implies that groups with high barriers to the outside will tend to
emphasize traditional customs and proper attention to the details of
ritual. This could be reflected in the stratum of the TSol transmitted on
the papyrus of chapter 18, which represents the oldest datable physical
evidence for any portion of what is now the TSol. This chapter relates
the story of thirty-six decans, or degrees of the zodiac, who appear very
briefly and describe themselves even more tersely and formulaically
than is typical for the rest of the collection. This section is relatively
unconcerned with any "higher" theological agenda or universal values,
and instead is singularly attentive to the precise names of power and
rituals of disposal. This might reflect the attitudes of people outside of
traditional avenues to power, such as nonelites (whether Jewish,
Christian, or "pagan") who experience negative events and tend to
blame demons for their problematic experiences in the world. People in
this social setting would likely be drawn to ritual solutions to solve oth
erwise insoluble problems—hence the emphasis on the efficacy of the
incantations provided here in Solomon's name. In this section, little
attention is paid to the narrative frame or historical biography of
Solomon, and there is no mention in the body of this chapter of the
Temple or Solomon's larger purpose in summoning the demons before
him. This is especially noteworthy because the usual pattern in the rest
of the collection has each demon consigned to Temple building labor
immediately after his or her introductory conversation with Solomon.

This content downloaded from 132.208.246.237 on Sun, 27 Jan 2019 20:50:31 UTC
All use subject to https://about.jstor.org/terms
 Schwarz: Demons and Douglas 923

If the introductory and concluding sections to the chapter19 are, in fact,

later efforts to incorporate an earlier independent document into the
collection, this further highlights the contrast between the demonologi
cal beliefs held by the composers of the decans material in chapter 18
over against those who integrated it into the emerging TSol collection.
The narrative framework which recounts the plot of the eventual
TSol collection, in contrast, spends more time concerned with "larger"
theological issues, God's plan, and the connection to the larger stream
of the tradition. This might reflect the concerns of people closer to the
center of power, such as priests or religious leaders, who are more con
nected to the mainstream as a way to resolve their problems (although
of course magical solutions and unconventional approaches seem
important to this layer of the text as well). There are also those ele
ments of the TSol which seem concerned with addressing the religion
of the other, and situating it firmly within the demonic camp. For
example, Onoskelis, the first female demon introduced to Solomon,
describes her role as a goddess tempting people to false worship and
evil ends. In describing her sphere of harm, she explains, "It is also true
that they worship my star secretly and openly. They do not know that
they deceive themselves and excite me to be an evildoer all the more.
For they want to obtain gold by remembering (me), but I grant little to
those who seriously worship me" (TSol 4.6-4.7, eclectic text). This
material might have been generated by people feeling beleaguered by
the world around them, in a closed community which saw all outsiders
as tainted. Thus, perhaps this demonization of "paganism" might have
originated in a sectarian moment, within which the individuals felt par
ticularly under attack from the outside (perhaps in a time of persecu
tion—of which there are many throughout the late antique and
Byzantine periods).
Finally, there is the question as to whether the final, composite TSol
can be attributed to a particular social location. The end products of
the collection process were generated by persons who seem to have
been able to hold seemingly competing cosmologies in their heads at
once, which seems to suggest something about social location.
Obviously, we know that this end product was generated by literate
individuals, although the manuscript evidence we have for the work of

1918.1-3, in which the entire group appears and speaks as one, calling themselves stoicheia,
rather than decans; and 18: 41-42, which seems to tie this material back to the narrative frame by
switching from the litany of demons each reporting their sphere and adjuring mechanism in turn,
to returning to Solomon, speaking in first person, glorifying God and returning to think of the
Temple.

This content downloaded from 132.208.246.237 on Sun, 27 Jan 2019 20:50:31 UTC
All use subject to https://about.jstor.org/terms
924 Journal of the American Academy of Religion

these editors and compilers suggests quite varied degrees of proficiency

with regard to standard scribal practices, the unique challenges of
copying spellbooks, and even the linguistic varieties of ancient versus
more modern Greek. While the challenges of dating this collection pre
cisely are clear, the manuscript evidence gives us reason to begin with
the presumption of a Byzantine date for the formulation of these fuller
manuscripts.
With these data in mind, it appears that these compilers might best
be slotted into the weak group/strong grid classification, because in this
setting people experience power as a remote and arbitrary force exer
cised by capricious rulers (whether human or divine). People in this
category view the hierarchy as relatively arbitrary, and thus would tend
to be less interested in reconciling the details of the demonic hierarchy
as presented in the TSol. Instead, people from a weak group/strong grid
setting might seek to widen the TSol collection, even if it resulted in
inconsistencies, so as to integrate as many efficacious sources of power
as possible because their main avenue to power is through that kind of
technique. One major objection to envisioning the TSol collectors in
this category would be their apparent access to relatively high-status
education, with the evident ability to write and transmit texts. Some
might suspect that such individuals, with access to status through edu
cation and the church, would copy such materials out of antiquarian
interest, rather than in quest of preservation of the power of such mate
rial (to the degree that they were uninterested in correcting its more
startlingly heterodox ideas)—and thus they would be better situated in
weak group/weak grid. However, we do find examples in the Byzantine
world of spellbooks copied, evidently for real use. For example, the
Kyranides, a spellbook and demonological collection like the TSol, has a
long and complex textual history.20 Kyranides was evidently consulted
and used in the Byzantine period, as witnessed by the condemnatory
reference in Patriarch Athanasius' letter (c. 1303), in which he wrote
that "corruption taints the [morally] weak and provokes many either to
see Koiranides or to entrust everything to these [books]" (Talbot 1975:
168). This confirms that at least this one example was not merely
copied as a book of curiosities, but had a real audience, willing to
"entrust everything" to it—or at least that individuals in the church
hierarchy were concerned that such a thing was plausible among literate
individuals under the purview of the patriarch's pen. The evidence that
Solomonic spellbooks were in continued circulation is further

20For the critical edition, see Kaimakis (1976).

This content downloaded from 132.208.246.237 on Sun, 27 Jan 2019 20:50:31 UTC
All use subject to https://about.jstor.org/terms
 Schwarz: Demons and Douglas 925

demonstrated by examples such as "Choniates' mention of the . . .

[Solomonic books] . . . found in the possession of Isaac Aaron in 1172,
which was designed to summon the demons in legions and make them
hurry to perform whatever task they were given."21 This is significant
because it confirms that Kyranidés was not the only demonological
compendium in circulation (Talbot 1975: 168, lines 80-81). While such
references do not prove that the books were used by their tradents,
rather than merely copied for antiquarian purposes, a patriarch's con
demnation of those who rely on such materials is suggestive.
Another relevant example which was current in Byzantine circles is
the work called Solomon's Magical Treatise.22 Richard Greenfield writes
that it "survives, mostly in fairly small fragments, in a number of manu
scripts of which the earliest date to the fifteenth century" (1988: 159).
He continues, "In its present basic form the Treatise is apparently a
fusion of various magical techniques, primarily for controlling demons"
(Greenfield 1988: 159-160). Most intriguingly,

[although it seems probable that the various astrological lists and such
like were at some time independent of the summoning rituals, the
obvious similarities between the main rite of the circle and the other
hydromancies and iatromancies which also appear here may argue
against this view and suggest that it is the basin or mirror that has
dropped out of the main ritual rather than the other rituals being
appended to something that originally had nothing to do with them.
(Greenfield 1988: 160)

This suggests a complex composite collection, in which the final

developed form reflects disjunctures in the flow of text which result
from its transmission history. His conclusion about the textual history
of the Treatise is: "At some point prior to the fifteenth century . . .
there will have been in existence a pseudo-Solomonic work which was
basically a hydromancy text-book. This will have had at its heart a
ritual for summoning demons to a circle by means of a water-basin . . .
but it will also have included alongside this instructions and detailed
information required for the complex preparations which preceded the
summoning, involving the use of power and concepts drawn from
magical and catarchic astrology" (Greenfield 1988: 161). This example

■"Greenfield (1995: 130), citing van Dieten (1975: 146, lines 45-47). This is noted by
Preisendanz (1956) and McCown (1922: 101-102).
22For publication of an important MS, see Delatte (1949). Also cf. the Hygromancy (sic) of
Solomon published by Torijano (2002).

This content downloaded from 132.208.246.237 on Sun, 27 Jan 2019 20:50:31 UTC
All use subject to https://about.jstor.org/terms
926 Journal of the American Academy of Religion

corresponds to the TSol not only in textual development, but also in

the surviving evidence (fragments and late manuscripts) as well as in
content (encyclopedia of magico-medical ideas with an astrological
flavor and a Solomonic theme).
These examples suggest further pieces of evidence which might help
situate the TSol and its theorized tradents, but are bolstered by consid
eration of what we know of the world from which they emerged. In
Greenfield's study of sorcery accusations in twelfth-century Byzantine
circles, he writes, "Sorcery being referred to in all these cases belongs to
the literary, learned tradition of Greek magic rather than some illiterate,
oral tradition of 'popular' Byzantine culture" (1993: 78). He also
observes that "belief in magic and sorcery was not a phenomenon con
fined to some sort of 'folk religion' which founds its followers only or
even chiefly in the uneducated and lower strata of Byzantine society"
(79), but instead cites several church authorities who confirm the power
of magic and sorcery even as they decry its use (80). Attention to this
medieval Byzantine context, then, confirms belief in the efficacy of
spellbooks even among relatively elite and powerful figures in the
church.
Another way of strengthening the suggestion that the tradents are
best understood within weak group/strong grid comes through consid
eration of categories of ritual expertise as studied by David
Frankfurter. Frankfurter describes local experts who "weave the caden
ces and mythology of orthodox liturgy and cosmology with the exi
gencies and spirits of the local cosmos" (2002: 168). He writes that
such individuals serve as intermediaries between the Great Tradition
and their local clientele, and do so by virtue of both their literacy and
their quasi-official status (2002: 169). Such individuals might derive
status from both their access to the official tradition and from their air
of otherness, "combining new and old idioms" (2002: 177) of power
and efficacy. Frankfurter's model suggest some conservatism might
operate among such individuals, who even as they innovated within
their ritual manuals might tend to preserve what came before as part
of the source of their expertise.
The end-stage TSol manuscripts present challenges to our typical
understanding of textual collections. These manuscripts shift in plot
and voice in ways that appear to disrupt the narrative while they
move through demonologies and cosmologies that seem mutually
incompatible. As they proceed, these late TSol collections transmit seem
ingly orthodox ideas about the power of Jesus, reverence for the Hebrew
Bible and New Testament, etc., even as they continue to include more
surprising and heterodox notions of power drawn from polytheist

This content downloaded from 132.208.246.237 on Sun, 27 Jan 2019 20:50:31 UTC
All use subject to https://about.jstor.org/terms
 Schwarz: Demons and Douglas 927

traditions and even unorthodox theologies which place the almighty God
as just another thwarting angel among many. In attempting to make
sense of this, I argue that Douglas' schema, supported by attention to
what we know of the historical context that might have produced these
end-stage collections, provide strong clues. By considering the historical
context of Byzantine spellbooks and taxonomies of ritual experts, it
seems possible to suggest that the ability to transmit such varied material
without "correction" is an important clue toward positing weak group/
strong grid as a likely location for the late tradents of this material.

CONCLUSION

In all, Douglas' work is highly useful for providing ways to think

about the varied demonologies present in the final TSol. Althou
there is some circularity in arguing that a fragmented text wa
brought together from once independent parts, and using Dougl
theory to argue that the parts I claim were once independent a
different, this effort moves closer, perhaps in spiral fashion,
extracting meaning from this challenging and complex text. Differe
types of demonologies appear to account for the differences am
strands of the TSol, and as a result of the presence of these competi
ideas within one text, I theorize that they come from different social
tions. With regard to the specific textual problems, this theory offer
useful explanatory lens. The full TSol contains diverse representations
demons, and these may reflect diverse demonologies associated with
ferent social locations. This proposal offers a possible way of accountin
for the amalgamation of the varying notions about demons in the te
and it makes it possible to make some suggestions about the hands th
textual components might have passed through over time. Furthermor
the notion that demonology tells us something meaningful about a soc
group is useful not just for the TSol, but for any text that includes signif
cant demonological information. The way deep fears about malevolen
forces are encoded and systematized indicates something about t
central concerns of a society. Thus, in different social settings, belie
about demons and how they should be handled will tend to look diff
ent. When demonic language is deployed, it speaks much more ab
the accuser than the accused, and thus, in accusations of demon worsh
we gain important hints about the individual or group making t
accusation.

This content downloaded from 132.208.246.237 on Sun, 27 Jan 2019 20:50:31 UTC
All use subject to https://about.jstor.org/terms
928 Journal of the American Academy of Religion

REFERENCES

Alexander, Phillip S."Contextualizing the Demonology of the

2003 Testament of Solomon." In Die Dämonen, ed.
A. Lange, Hermann Lichtenberger and
K. F. Diethard Romheld, 613-635. Tübingen,
Germany: Mohr Siebeck.

Bell, Catherine M. Ritual Theory, Ritual Practice. New York, NY:

1992 Oxford University Press.
1997 Ritual: Perspectives and Dimensions. New York,
NY: Oxford University Press.

Brakke, David Demons and the Making of the Monk: Spiritual

2006 Combat in Early Christianity. Cambridge, MA:
Harvard University Press.

Brown, Peter "Sorcery, Demons, and the Rise of Christianity."

1970 In Witchcraft Confessions and Accusations, ed.
M. Douglas, 17-46. London, UK: Tavistock
Publications.

1978 The Making of Late Antiquity. Cambridge,

Harvard University Press.

Daniel, Robert "The Testament of Solomon XVIII 27-28,

1983 33-40." In Papyrus Erzherzog Rainer (P. Rainer
Cent.): Festschrift zum 100-jährigen bestehen der
Papyrussammlung der österreichischen national
bibliothek. Vol. 1; 1294-1304. Wien, Austria:
Hollinek.

Delatte, A. "Le traité des plantes planétaires d'un manu

1949 scrit de Léningrad." Annuaire De l'Institut De
Philologie Et d'Histoire Orientales Et Slaves 9:
145-177.

Douglas, Mary Cultural Bias. London, UK: Royal

1978 Anthropological Institute, occasional paper
no. 35.

1982a "Introduction to Grid/Group Analysis." In Essays

in the Sociology of Perception, ed. M. Douglas,
1-8. London, UK, Routledge and Kegan Paul.
1982b Natural Symbols: Explorations in Cosmology.
New York, NY: Pantheon Books.

This content downloaded from 132.208.246.237 on Sun, 27 Jan 2019 20:50:31 UTC
All use subject to https://about.jstor.org/terms
 Schwarz: Demons and Douglas 929

Duling Dennis "The

C. Testament of Solomon." In The Old
1983 Testament Pseudepigrapha, ed. J. H. Charles
worth. Vol. 1; 1935-1987. Garden City, NY:
Doubleday.

Dennis C. Duling "The Legend of Solomon the Magician in

1984 Antiquity: Problems and Perspectives."
Proceedings of the Eastern Great Lakes and
Midwest Biblical Society 4; 1-22.
1988 "The Testament of Solomon: Retrospect and
Prospect." Journal for the Study of the
Pseudepigrapha 2:87-112.

Frankfurter, David "Dynamics of Ritual Expertise in Antiquity and

2002 Beyond; Towards a new Taxonomy of "Magicians'."
In Magic and Ritual in the Ancient World, ed.
P. A. Mirecki and M. W. Meyer. Leiden, The
Netherlands: Brill.

2006 Evil Incarnate: Demonic Conspiracy and Ritual

Abuse in History. Princeton, NJ: Princeton
University Press.

Garcia Martinez, F. The Dead Sea Scrolls Translated: The Qumran

1996 Texts in English. Leiden, The Netherlands,
New York, NY, and Grand Rapids, MI: E.J. Brill
and W.B. Eerdmans Pub.

Greenfield, Richard P. H. Traditions of Belief in Late Byzantine Demonology.

1988 Amsterdam, The Netherlands: Adolf M. Hakkert.

1993 "Sorcery and Politics at the Byzantine Court

in the Twelfth Century: Interpretation of
History." In The Making of Byzantine History:
Studies Dedicated to Donald M. Nicol, ed.
Roderick Beaton and Charlotte Roueche, 73-85.
Aldershot, UK: Variorum.

1995 "A Contribution to the Study of Palaeologan

Magic." In Byzantine Magic, ed. H. Maguire,
117-153. Washington, DC: Harvard University
Press.

This content downloaded from 132.208.246.237 on Sun, 27 Jan 2019 20:50:31 UTC
All use subject to https://about.jstor.org/terms
930 Journal of the American Academy of Religion

Johnston, Sarah Iles "The Testament of Solomon from Late

2002 Antiquity to the Renaissance." In The
Metamorphosis of Magic from Late Antiquity to
the Early Modern Period, ed. J. N. Bremmer
and J. R. Veenstra, 35-49. Leuven, Belgium:
Peeters.

Josephus, F. and The Works of Josephus. Peabody, MA:

W. Whiston Hendrickson Publishers.
1987

Kaimakis, Dimitris Die Kyraniden. Meisenheim am Glan, Germany:

1976 Verlag Anton Hain.

Klutz, Todd Rewriting the Testament of Solomon: Tradition,

2005 Conflict and Identity in a Late Antique
Pseudepigraphon. London, UK and New York,
NY: T&T Clark.

Roy Kotansky "Incantations and Prayers for Salvation on

1991 Inscribed Greek Amulets." In Magika Hiera:
Ancient Greek Magic and Religion, ed.
C. A. Faraone and D. Obbink, 107-117.
New York, NY: Oxford University Press.

Kraemer, Ross S. Her Share of the Blessings: Women's Religions

1992 among Pagans, Jews, and Christians in the
Greco-Roman World. New York, NY: Oxford
University Press.

McCown, Chester C. The Testament of Solomon. Leipzig, Germany:

1922 J.C. Hinrichs.

1923 "The Ephesia Grammata in Popular Belief."

Transactions and Proceedings of the American
Philological Association LIV:128-140.

Preisendanz, Karl "Ein Wiener Papyrusfragment

1956 Testamentum Salomonis." EOS 48:161-7.

Preisendanz Karl "Salman." In Paulys Realencylopädie der Classis

1959 chen Altertumswissenschaft, ed. A. F. Pauly,
660-704. Munich, Germany: G. Wissowa.

Robinson, James M. The Nag Hammadi Library. New York, NY:

1990 HarperSanFrancisco.

This content downloaded from 132.208.246.237 on Sun, 27 Jan 2019 20:50:31 UTC
All use subject to https://about.jstor.org/terms
 Schwarz: Demons and Douglas 931

Schwarz, Sarah "Reconsidering

L. the Testament of Solomon
2007 Journal for the Study of the Pseudepigrapha
16/3:203-237.

Smith, J. Z. "Towards Interpreting the Demonic Powers in

1978 Hellenistic and Roman Antiquity." Aufstieg und
Niedergang der romischen Welt 11.16.1:425-439.

Spickard, James V. "A Guide to Mary Douglas's Three Versions of

1989 Grid/Group Theory." Sociological Analysis 50/2:
151-70.

Stewart, Charles Demons and the Devil: Moral Imagination in

1991 Modem Greek Culture. Princeton, NJ:
Princeton University Press.

Talbot, A. M. M. The Correspondence of Athanasius I, Patriarch

1975 of Constantinople. Washington, DC: distributed
by Harvard University Press.

Torijano, P. A. Solomon, the Esoteric King: From King to

2002 Magus, Development of a Tradition. Leiden,
The Netherlands: E.J. Brill.

van Dieten, Jan-Louis Nicetae Choniatae Historia. Corpus fontium his

1975 toriae Byzantinae. Vol. 11, 1-2. Berlin,
Germany: Berolini et Novi Eboraci: de Gruyter.

Nordheim, E. "Das Testament Salomos." In Die lehre der

1980 alten. Vol. 1; 1185-1193. Leiden, The
Netherlands: E.J. Brill.

This content downloaded from 132.208.246.237 on Sun, 27 Jan 2019 20:50:31 UTC
All use subject to https://about.jstor.org/terms