Mary Douglas’s Holiness/Wholeness Paradigm: Its Potential for Insight and its Limitations
Saul M. Olyan
Brown University


The strengths and weaknesses of Mary Douglas’s holiness/wholeness paradigm, in which she links the idea of holiness directly to physical wholeness, are critically assessed in this article. Several examples of fruitful elaboration of the paradigm by biblical scholars are considered in some detail. In addition, the author suggests some ways in which the paradigm might be modified in order to explain better the biblical data concerning physical wholeness.

In her influential essay “The Abominations of Leviticus,” published in 1966 in the volume Purity and Danger, Mary Douglas introduced what I have chosen to call the holiness/wholeness paradigm, in which she links the idea of holiness directly with physical wholeness or completeness.1 Though criticized in its details, the paradigm has been profitably elaborated and modified by biblical scholars, and core aspects of it remain influential.2 It is my purpose in this paper to explore the paradigm’s potential for insight as well as its limitations, and to suggest some ways in which it might be reconfigured in order to explain better the biblical data concerning physical wholeness.

In her exploration of the concept of holiness in “The Abominations of Leviticus,” Douglas noted the biblical emphasis on wholeness and completeness, and linked these directly to the holy. For Douglas, “holiness is exemplified by completeness”; in fact, “the idea of holiness was given an external, physical expression in the wholeness of the body seen as a perfect container.” “To be holy is,” for Douglas, “to be whole.”3 In a later essay, “Deciphering a Meal,” Douglas reiterates this association of holiness with wholeness in a slightly different way: “The sanctity of cognitive boundaries is made known by valuing the integrity of the physical forms.”4 In both “The Abominations of Leviticus” and “Deciphering a Meal,” Douglas understands wholeness as an articulation of holiness: it is an “external, physical expression” of it; it exemplifies it; it makes it known. In short, wholeness is understood as a communicator of holiness. In her more recent work Leviticus as Literature, Douglas reiterates the core ideas of the paradigm: “Only the perfect body is fit to be consecrated, no animal with a blemish may be sacrificed, no priest with a blemished body shall approach the altar.…”5

Certainly there is more than a little evidence to support a linkage between the holy and the whole in the biblical text. Nearly all sacrificial animals presented before Yhwh had to be “without (physical) ‘defect’ ” (kol-mûm lô’ yihyeh-bô) or “whole” (tāmîm) according to Lev 22:17–24.6 That most if not all of these sacrifices were sanctified is suggested by a variety of data, including the common label “holy things” or “holy foods” (qodāšîm) used of offerings reserved for the priests and, in some cases, their dependents, and by characterizations such as “the holy foods which the children of Israel have sanctified to Yhwh” (Lev 22:3). Other texts suggest that offerings not formally classed as qodāšîm were also sanctified. Lev 19:8 (H) states clearly that the well being offering (šĕlāmîm) is holy, and Lev 7:19–21 (P), which restricts the eating of the meat of the well being offering only to clean persons, and threatens those who would violate this restriction with termination of lineage (kārēt), suggests as much.7 The link between holiness and wholeness is also evident in Deut 15:19–23, which commands the sanctification of first-born male sacrificial animals to Yhwh. The exceptions to the rule of sanctification are those male cattle, sheep and goats with a “defect” (mûm). According to this text, such animals are not to be brought to the sanctuary and sacrificed; instead, they are to be treated as non-sacrificial game animals are treated, eaten in settlements by the unclean and clean alike after their blood is removed. The fact that persons who are unclean may eat sacrificial animals with “defects” indicates clearly that they are not understood to be holy, for that which is holy must be guarded from pollution, and the text permits unclean persons to eat defective animals. Thus, according to Deut 15:19–23, first-born male sacrificial animals with a “defect” remain unsanctified because they are not whole. Holiness and somatic wholeness are also related in Lev 21:17–23, which requires priests who offer sacrifices to Yhwh and priests of the high priestly line who approach the curtain of the holy of holies to be “defect”-free (= whole): “But to the curtain he shall not come, nor shall he approach the altar, for he has a “defect” (mûm); he shall not profane (wĕlôʾ yĕḥallēl) my sanctuaries, for I, Yhwh, sanctify them” (v. 23). That priests are holy according to Leviticus 21 is indicated in v. 7 (“For holy is he [the priest] to his god”) and in v. 8 (“You shall treat him as sacred, for the food of your god he brings near …”).

As has been pointed out, Douglas, who also spoke of sacrificial animals and priests in her treatments, was not correct to claim that all sacrificial animals, all priests and all worshipers had to be physically whole, without “defect,” to gain access to the sanctuary.8 Lev 22:23 allows the sacrifice of animals with two specific “defects” (probably limbs of uneven length) as free-will offerings;9 Lev 21:22 permits priests with “defects” to remain in the temple and eat holy and most holy foods;10 Deut 23:2 likely forbids men with genital damage from entering the temple, and 2 Sam 5:8b may bear witness to a proscription of worshipers with “defects,” but no text clearly bans all blemished worshipers, and—interestingly—no text in the P/H tradition even hints at any prohibition of worshipers with “defects.”11 Douglas was also incorrect to suggest—at least as I read her—that sacrifices, priests, worshipers and soldiers are all constructed as holy by biblical sources.12 Certainly priests are hallowed, as are most if not all sacrifices, but according to P, the possession of sanctity distinguishes priests from all other Israelites, including worshipers and soldiers (e.g., Exod 29:33; Num 16:1–17:5).13 Finally, Douglas tends to blur the biblical distinction between a lack of somatic wholeness (i.e., having a “defect” [mûm]), and impurity. This is clear in her classification of the “leper” and the parturient, both polluters, with persons and sacrificial animals that have “defects” (mûmîm). Though the “leper” and parturient are unclean, their pollution does not render them “defective” (= not whole), and therefore, they ought not to have been included in Douglas’s discussion. The same is true of persons with bodily discharges and priests polluted by corpse contact, both mentioned by Douglas among those who are not whole.14 One might add that unclean animals also have nothing necessarily to do with that which is defective, and clean animals are holy only when designated for sacrifice, if then.15 Even given these weaknesses in Douglas’s formulation, the link she established between holiness and physical wholeness is nonetheless evidenced, though not to the degree and with the consistency that she claimed.

The holiness/wholeness paradigm has been elaborated in recent years by a number of biblical scholars in ways that suggest its continued utility. I shall speak of three specific examples of its elaboration, in order of their appearance in the scholarly literature.16 In a 1996 article published in ZAW, I argued that Douglas’s holiness/wholeness paradigm is evidenced in biblical materials even more extensively than she had suggested.17 My focus was the stones of the altar in Exod 20:25 and Deut 27:5–6, as well as the stones of the temple in 1 Kgs 6:7. Exod 20:25 forbids an altar made of ashlar (cut stone), warning that working altar stones with a tool profanes them: “If you wield your tool upon it, you profane it (wattĕḥalĕlehā).”18 This statement indicates that according to Exod 20:25, altar stones, like most sacrifices and like priests, are sanctified. If this were not the case, the stones would not be subject to profanation. (Profanation transforms that which is holy into that which is common.) Deut 27:5–6, elaborating Exod 20:25, also forbids the use of a tool (explicitly iron) on the stones; it refers to the uncut stones from which the altar is to be built as “whole stones” (ʾăbānîm šĕlēmôt). Thus, the unworked “whole stones” of Deut 27:6 parallel the uncut holy stones of Exod 20:25. This suggests a connection between the wholeness of the uncut altar stones and their holiness, which is lost according to Exod 20:25 if they are worked with a tool. If I am correct about this connection, then we can compare Deut 15:21. Just as male first-born sacrificial animals with a “defect” are not sanctified according to Deut 15:21, so altar stones that lose their wholeness lose their holiness. In both instances, that which is whole is understood to be holy, and that which lacks wholeness is treated as common. According to 1 Kgs 6:7, the temple, like the altar of Deut 27:6, was to be constructed from “whole stone” (ʾeben šĕlēmā). The verse alludes directly to Deut 27:5–6 by mentioning “whole stone” and noting the absence of iron tools when the temple was built. Yet 1 Kgs 6:7 concerns the building of the temple, not the erection of the altar. It apparently applies the altar law to the construction of the temple, thereby elaborating it. Is the whole stone used to erect the temple holy? Though not stated explicitly, the whole stone may well be assumed to be sanctified by the text, given that the altar stones in the same D tradition appear to be (Deut 27:5–6, elaborating Exod 20:25), and given that other, non-D texts (e.g., P) understand the various implements of the sanctuary complex to be holy (e.g., Exod 30:22–29).

A second, recent elaboration of the holiness/wholeness paradigm is Jacob Milgrom’s notion of “blemished time,” introduced in his analysis of the festivals of Leviticus 23.19 Building explicitly upon my treatment of the altar stones, and implicitly on Douglas’s original articulation of the holiness/wholeness paradigm, Milgrom argues for an analogy between sacred items and sacred time: “Just as the altar must be whole, so must sacred time. As human activity with stone desecrates the altar, so does human activity in time: work. Both space and time in their holy dimension must remain in their natural state; they may not be blemished or desecrated by human labor. To be sure, blemished time is an abstraction. It is not visible, as are blemished space and the changed appearance of a blemished priest, sacrificial animal, or altar …”20 Milgrom understands the Sabbath (hallowed time) as metaphorically whole and subject to “blemish” and desecration via human labor just as holy items such as altar stones are subject to a loss of wholeness and sacredness through being worked with a tool.

Susan Niditch’s use of the paradigm to illuminate the proscription of priestly hair cutting and high priestly hair manipulation of any sort in mourning contexts is a third, and very recent, example of elaboration.21 Why must priests eschew shaving rites and high priests the manipulation of intact hair after the death of a close relative? The answer given by the text, as Niditch points out, is priestly holiness.22 Citing Douglas’s point that holiness is related to somatic wholeness, Niditch argues that “the priest needs to maintain bodily boundaries demarcated by intact hair and body.” Thus, she elaborates upon Douglas’s paradigm, arguing that holiness requires wholeness of hair as well as that of body. (Douglas had mentioned the bodies of priests in relation to “defects” [mûmîm], but had said nothing about hair manipulation.) Niditch further develops her elaboration of the holiness/wholeness paradigm in her discussion of restrictions on the high priest. “The high priest is even more holy than the priest, and thus his relation to death and to the manipulation of hair in connection with death is even more circumscribed.” For the high priest, even disheveled hair “interrupt[s] his wholeness and holy status.”23 Whether or not one finds Niditch’s explanation of proscriptions on priestly and high priestly hair manipulation, Milgrom’s notion of “blemished time,” or my treatment of the stones of the altar and temple convincing, they illustrate well the impact of Douglas’s thinking on biblical scholarship as well as the potential utility of Douglas’s holiness/wholeness paradigm for explaining phenomena of the cult.

What of the paradigm’s limitations? Douglas argued that wholeness expresses holiness in a physical way, that the valuing of integrity communicates sanctity, that “holiness is exemplified by completeness.” If I understand Douglas correctly, wholeness acquires its significance through its relationship to holiness, as an expression or embodiment of it. Thus, for Douglas, the privileging of wholeness is the result of its relationship to holiness. This view of wholeness strikes me as overly limited in scope, for integrity of form can be shown to be prioritized in the world of the biblical text apart from considerations of holiness. In short, wholeness is desirable even in cases where holiness is not at issue. A prime example of this is the relationship between wholeness and beauty in a number of biblical texts. In each of the following examples, male or female beauty is discussed in a context completely removed from considerations of holiness. According to 2 Sam 14:25, Absalom’s beauty could not be matched in all Israel. He is described as “a handsome man” (’îš yāpeh) and the text goes on to state that “from the bottom of his foot to the crown of his head, there was no ‘defect’ (mûm) in him.” Thus, for the author of this text, Absalom’s wholeness, indicated by his complete lack of physical “defects,” is emblematic of his beauty. The same notion that wholeness is emblematic of beauty is to be found in the Song of Songs. In 4:7, the male lover describes the appearance of his beloved: “You are completely beautiful, my companion,” // “There is no ‘defect’ (mûm) in you.” In 6:9, the female lover, praised as singular, is described as “my perfect one,” tammātî, an adjective treated as a substantive which is derived from the same root (tmm) as tāmîm/tĕmîmā, “whole,” the antithesis of defective; it is likely a reference to the female lover’s physical appearance.24 Finally, Dan 1:4, listing the attributes of the Judean youths to be recruited to Nebuchadnezzar’s court, includes both beauty of appearance (ṭôbê mar’eh) and a lack of “defects” (’ên bāhem kol-mûm). In all of these examples, wholeness is closely associated with beauty; in several, it is not only characteristic of the beautiful, it is emblematic of it. Douglas’s claim that “holiness is exemplified by completeness” could also be made about beauty as it is presented in texts such as 2 Sam 14:25 and Song 4:7. That which is whole is not necessarily understood to be beautiful or holy, but wholeness is not infrequently emblematic of both holiness and beauty.25 Given that the holiness/wholeness paradigm as formulated by Douglas does not account for the prioritization of wholeness apart from considerations of holiness, I propose to reformulate it with the focus shifted from holiness to wholeness: Physical wholeness may exemplify beauty or holiness.26 With this change of focus, the paradigm does a better job explaining the evidence of the text. Douglas’s correct observation that wholeness exemplifies, expresses, makes known is preserved; her overly narrow notion of what wholeness communicates is jettisoned.

The exceptions not accounted for by Douglas’s holiness/wholeness paradigm also suggest its limitations as an explanatory tool. As mentioned, the paradigm does not explain exceptional holy persons and animals that are not whole. Though a priest with a “defect” (mûm) may not approach the altar, and a potential high priest with a “defect” may not approach the curtain of the holy of holies, according to Lev 21:22, priests with “defects” retain access to the sanctuary and to holy items such as the holy and most holy foods: “The food of his god, from the most holy foods and from the holy foods, he may eat.” There is no hint in this text that the priest with a “defect” loses his sanctity, or loses access to all things holy. Quite the opposite. Likewise, according to Lev 22:23, sacrificial animals with certain, specific “defects” (likely limbs of uneven length) may be sacrificed to Yhwh as freewill offerings.27 Though some biblical texts that I have discussed suggest that a loss of wholeness results in a loss of holiness (e.g., Exod 20:25; Deut 27:5–6; Deut 15:21), or in loss of access to some holy space or holy items (Lev 21:23), this is not the case in every instance. How then are these exceptions to be explained? Though the priest with a “defect” does not lose his sanctification or his access to holy foods according to H, he does lose his access to the primary priestly activity: offering sacrifice. And though Lev 22:23 permits the offering of sacrificial animals with two particular “defects” as freewill offerings, it does not allow such animals to be presented to fulfill vows (or, by implication, as thanksgiving offerings). In other words, defective priests cannot perform central, elite rites, nor can defective sacrificial animals be presented as higher-ranked offerings according to Leviticus 21 and 22 (H). Though tolerated, these exceptional cases nonetheless point to the inferiority of that which is defective and the desirability of that which is whole. The exceptions, whatever their motivation, reiterate the ideal of wholeness as an attribute of the holy, but the fact that there are exceptions suggests that Douglas’s absolute claims (e.g., “To be holy is to be whole”) limit her paradigm’s utility as an explanatory tool. It is also important to point out that Douglas’s paradigm may be better supported by some biblical sources than by others, as the exceptions I have noted are both present only in H texts. When we look at D and the Book of the Covenant, no exceptions challenge the requirement of wholeness for holiness (Exod 20:25; Deut 27:5–6; 15:21). Though Douglas did not distinguish between biblical sources, it is important that we do so if we are to evaluate the utility of her paradigm with any insight.

[1] “The Abominations of Leviticus,” in Purity and Danger: An Analysis of the Concepts of Pollution and Taboo (1966; London: Ark, 1984), 41–57. Though Douglas focuses mostly on physical forms of wholeness, she also includes non-somatic examples in her discussion. On these, see further n. 26.

[2] For two important examples of the influence of core aspects of the paradigm, see, e.g., J. Milgrom, Leviticus 1–16 (New York: Doubleday, 1991), 721: “To be sure, her definition of the term ‘Holy as wholeness and completeness’ … is justified …”; G. J. Wenham, The Book of Leviticus (Grand Rapids, MI: Eerdmans, 1979), 23–25, 169: “In our Introduction[] it was suggested that the notion underlying holiness and cleanness was wholeness and normality.” See also P. J. Budd, “Holiness and cult,” in R. E. Clements, ed., The World of Ancient Israel: Sociological, Anthropological and Political Perspectives (Cambridge: Cambridge University Press, 1989), 286–90 for synopsis and analysis of reactions to Douglas’s ideas about wholeness and holiness, mainly in relation to the dietary laws. Especially notable in this regard is M. P. Carroll’s critique and reapplication using Levi-Strauss’s Nature/Culture binary (“One More Time: Leviticus Revisited,” Archives européennes de sociologie 99 [1978]: 339–46). See also E. Leach, “Anthropological Approaches to the study of the Bible during the twentieth century,” in Structuralist Interpretations of Biblical Myth, ed. E. Leach and D. A. Aycock (Cambridge: Cambridge University Press, 1983), 20–21, and J. Duhaime, “Lois alimentaires et pureté corporelle dans le Lévitique. L’approche de Mary Douglas et sa reception par Jacob Milgrom,” Religiologiques 17 (1998): 19–35. On the dietary laws and the holiness/wholeness paradigm, see my discussion ahead, and in n. .

[3] “Abominations,” 51–52, 53, 54.

[4] “Deciphering a Meal,” Daedalus 101 (1972):61–81; 76–77 for the quotation. Here, she emphasizes the value placed on wholeness.

[5] Leviticus as Literature (Oxford: Oxford University Press, 1999), 46.

[6] I shall discuss the few exceptions ahead.

[7] To Lev 7:19–21 one might compare Lev 22:3 (H), which uses very similar wording to speak of the punishment of those priests who have contact with a holy food while unclean. If the well being offering’s meat were not sanctified according to P, there would be no need to regulate the purity status of those who have contact with it (cf. Deut 15:22).

[8] E.g., “Abominations,” 51: “Much of Leviticus is taken up with stating the physical perfection that is required of things presented in the temple and of persons approaching it.” For a detailed critique of Douglas on this matter, see, e.g., Milgrom, Leviticus 1–16, 720–21. Milgrom’s “reception” of Douglas is discussed at length by Duhaime, “Lois alimentaires.”

[9] On the specific “defects” in question (úârûaʿ, qâlûṭ), see the discussion of Milgrom, Leviticus 17–22 (New York: Doubleday, 2000), 1078. H, the group responsible for Lev 22:23, appears to rank freewill offerings lower than vows and, by implication, thanksgiving offerings, given that animals with these “defects” are only acceptable as freewill offerings. Compare P, which apparently ranks the thanksgiving offering above the vow and the freewill offering (Lev 7:15–16). Milgrom’s critique of Douglas misses the fact that Lev 22:23 allows this exception regarding defective sacrificial animals: “The altar … is served only by whole (unblemished) animals and priests …” (Leviticus 1–16, 721).

[10] This point is also noted by Milgrom, Leviticus 1–16, 721. For the holy and most holy foods, offerings reserved for the priests and their dependents, see, besides Lev 21:22, Lev 22:2–7, 10–13 and Num 18:8–19.

[11] On Deut 23:2 and its probable reference to entering the sanctuary, and on the adage of 2 Sam 5:8b, see my discussion in Rites and Rank: Hierarchy in Biblical Representations of Cult (Princeton: Princeton University Press, 2000), 107–111.

[12] See, e.g., “Abominations,” 51, regarding soldiers: “The army could not win without the blessing and to keep the blessing in the camp they had to be specially holy.” The notion that worshipers are holy is implicit in Douglas’s treatment.

[13] On Num 16:1–17:5 as a P text, see Olyan, Rites and Rank, 136 n. 63. On H’s view of the sanctity of Israel (contrast P), see Olyan, ibid., 121–22.

[14] “Abominations,” 51.

[15] As is well known, Douglas analyzes the biblical dietary laws in the context of her development of the holiness/wholeness paradigm (“Abominations,” 54–57). It is noteworthy that this particular aspect of her treatment has elicited such a spirited and often positive response from biblical scholars, given its problematic relationship to biblical discourses on wholeness and holiness.

[16] Other examples of elaboration or modification could be discussed. See, e.g., the papers of P. Budd and M. P. Carroll cited in n. 2. I chose the three examples I discuss because they illustrate well a number of the ways in which Douglas’s paradigm might be supported through elaboration. My critique of her formulation follows.

[17] “Why an Altar of Unfinished Stones? Some Thoughts on Ex 20, 25 and Dtn 27,5–6,” ZAW 108 (1996):161–71.

[18] Here I translate ḥereb as “tool” rather than sword, given the context.

[19] Leviticus 23–27 (New York: Doubleday, 2000), 1978–9.

[20] Ibid.

[21] “My Brother Esau is a Hairy Man”: Hair and Identity in Ancient Israel (New York: Oxford University Press, 2008), 106–107. The prohibitions in question are to be found in Lev 21:5 and 10.

[22] Niditch cites Lev 21:6 on the priest’s sanctification (ibid., 106); she could also have cited 21:15 on that of the high priest.

[23] Ibid., 107.

[24] Song 5:2 also has tammâtî used in reference to the female lover.

[25] There are countless examples of whole persons, animals or things that are neither holy nor beautiful according to our texts.

[26] There are other examples of the valuing of wholeness and completeness apart from considerations of holiness, but these tend to be non-somatic (e.g., Gen 6:9; 17:1, which use tâmîm in a behavioral sense: “innocent,” “having integrity”). Douglas included non-physical examples of wholeness in her original formulation (e.g., “rectitude and straight dealing”), though she associated them incorrectly with holiness, which she tended to distribute too liberally (“Abominations,” 52–53). Are there examples of wholeness exemplifying both beauty and holiness at the same time? Though no explicit examples of this exist to my knowledge, it may be implied by the evident association of defective, first-born sacrificial animals with both non-sanctification and ugliness in Deut 15:19–23. It is clear that according to this text, the whole, first-born sacrificial animal is to be sanctified; what remains unclear is whether it is also considered beautiful.

[27] That these are sanctified is likely given the witness of Lev 19:8 (H), which treats the well being offering as holy and subject to profanation, and the witness of Lev 22:21 (H), which classifies the free will offering as a type of well being offering (as does P: Lev 7:11–21).