1. Introduction
There is a long history of scholarship immediately declaring the problems with πυγμῇ in Mark 7.3 (ἐὰν μὴ πυγμῇ νίψωνται τὰς χεῖρας), with numerous commentators calling this NT hapax legomenon ‘mysterious’, ‘perplexing’, ‘elusive’, ‘obscure’ or even ‘nonsensical’, and with explanations regularly deemed unsatisfactory.Footnote 1 This scholarly puzzlement already suggests that we should be rejecting the textual alternatives to πυγμῇ in Mark 7.3 for related reasons: scribes then, as exegetes now, struggled with its meaning. While πυγμῇ has strong support (A, B, [D (πύκμῃ)], L, Θ etc.), there is the not insignificant variant πυκνά (e.g. א, W) and even the omission of πυγμῇ altogether (Δ, sys, sa). However, the simplest and most obvious explanation is that πυγμῇ is original and was replaced by ‘often’ or just dropped because Christian scribes did not understand the practice. We can see similar things happening elsewhere in Mark 7.3–4 where scribes struggled to understand the complexities of purity law outlined by Mark. As Joel Marcus notes, we have a choice between the Pharisees immersing themselves (βαπτίσωνται) and their cups, pitchers, copper vessels and couches (καὶ κλινῶν) or the Pharisees sprinkling themselves (ῥαντίσωνται) and immersing their cups, pitchers and coppers vessels.Footnote 2 The former represents known cultural practices, while the latter looks as if it is the result of attempts by later scribes to make puzzling Jewish purity practices intelligible.Footnote 3 It would seem that πυγμῇ caused similar cultural problems as βαπτίσωνται and κλινῶν.
In this article it will be argued that πυγμῇ in Mark 7.3 should not be considered mysterious as much of the history of scholarship would have it. It will be argued that πυγμῇ reflects a known purity practice concerned with using the minimal amount of water required when holding out the hand in the shape of a fist with water poured on it, and that this was the ideal pragmatic solution. As we will see, the fist would have been assumed to have been sufficiently relaxed in order that an economic use of water could seep through the fingers to cover enough of the hand required for proper purification. The literal translation ‘fist’, at least with the sense ‘in the form of a fist’, is therefore fully justified. We can then move on to see how Mark 7.3 is significant for understanding Mark 7.1–23 as a whole. Before all this, alternative suggestions need to be analysed.
2. Possible Solutions
Various possible solutions to the apparently mysterious πυγμῇ have, of course, been put forward, the most influential of which include: ‘a fistful (of water)’; ‘cupped hand’; or a measurement (up to the knuckles, wrist or elbow).Footnote 4 As we will see, there is certainly some concern for the measurement of water required for hand-washing in purity law (e.g. m. Yad. 1.1), but it reads too much into πυγμῇ and distorts its most obvious and literal meaning to render it ‘fistful’.Footnote 5 Furthermore, if ‘fistful’ were also to be read in terms of pouring water from cupped hands, we then not only face the semantic and exegetical problems but also a major cultural problem given that m. Yad. 1.2 claims that such a practice is not permitted: ‘With all sorts of utensils do they pour [water] for hands… Nor should a man pour [water] for his fellow with cupped hands’. To counter such potential criticisms, Martin Hengel suggested that we could be dealing with a disputed practice underlying m. Yad. 1.2.Footnote 6 However, against Hengel, there is no mention of alternative practices in m. Yad. 1.2, and it is worth adding that the context in m. Yad. 1.2 is non-polemical. Indeed, discussions of hand-washing elsewhere in m. Yadayim, and the Mishnah as a whole, certainly do note alternative interpretations and counter-arguments which further undermines the idea that the practice outlined in m. Yad. 1.2 is a disputed practice. We might add that Mark's handling of hand-washing and related practices in Mark 7.1–23 precisely matches the mainstream hand-washing practices in rabbinic literature so we might tentatively add that he would also be doing so once again with πυγμῇ.Footnote 7 Hengel's argument is also based on πυγμῇ effectively being a Latinism, which, he argued, was deliberately employed to make the meaning clearer to the audience.Footnote 8 More specifically, Hengel turned to pugillo which can refer to a ‘fistful’, with pugillo and pugnus being interchangeable in Latin. While πυγμή usually means clenched fist or a length, and not a ‘fistful’, Hengel looked to the OL which translates Mark 7.3 in the sense of a ‘handful’ (nisi pugillo laverint). Yet Hengel had already undermined his own argument because the Septuagintal equivalent of pugillus, δράξ (‘handful’; Lev 2.1–2; 1 Kings 17.12; Isa 40.12; Ezra 13.19), meant that Mark could simply have used the obvious word δράξ but did not.Footnote 9
The suggestion that πυγμῇ refers to a measurement of some part of the hand and/or forearm has some grammatical support, even if it has its own difficulties.Footnote 10 Occasional reference is made in the scholarly literature to Theophylact's eleventh century commentary on Mark which gives the following definition of πυγμῇ: ‘For it is not written in the law to wash πυγμῇ, that is, up to the elbow (for πυγμῇ means the part from the elbow to the tips of the fingers)’.Footnote 11 This reading has some support from the second-century CE grammarian, Julius Pollux, where πυγμή is from the elbow to the knuckles (Onomasticon 2.147, 158).Footnote 12 But these readings are too culturally removed from Jewish hand-washing practice as m. Yad. 2.3 requires washing to the ‘joint’ or ‘division’ (פרק) which does not seem to be as far as the elbow but rather knuckles or possibly wrist: ‘The hands are susceptible to uncleanness and are rendered clean up to the פרק’.Footnote 13 While from much later texts, it is worth noting that y. Ber. 8.2, 12a and b. Hullin 106a–b confirm this cultural practice. The shorter length (knuckles or possibly wrist) would also be consistent with the quarter-log (= one-and-a-half eggs) required for hand-washing in m. Yad. 1.1.
Developing this scholarly tradition, and after a typically learned discussion of purity issues, Roger Booth suggested that ‘when hands are washed in an ordinary non-cultic way to-day, the fingers and knuckles of the one hand are naturally rubbed in the palm of the other’ and placed this in line with m. Yad. 2.3 (‘The hands are susceptible to uncleanness and are rendered clean up to the joint/knuckle/wrist (פרק). How so? [If] one poured the first [water] up to the wrist, and the second beyond the wrist and it went back to the hand—it is clean…’).Footnote 14 However, there is a significant problem: there is no mention of ‘fist’ in m. Yad. 2.3 and if the hand-washing motion is so ‘natural’ then why was πυγμῇ included, a word which, we should not forget, confused interpreters from early on?
We will return to issues of the ‘joint’ (פרק) in due course but before that we need to turn to probably the most influential interpretation of πυγμῇ, that of Stephen Reynolds. Reynolds comes close to reading πυγμῇ literally (i.e. ‘with the fist’) but qualifies this by suggesting that hands would have been cupped. This reading was based on Reynolds’ own conversations with rabbis and his observations of modern Jewish practices. On these bases he made the following claims:
I believe that in NT times as today the hands were held with fingers flexed or cupped, so that they were neither tightly clenched fists nor open or spread wide. The purpose of not clenching the fist is to allow the water to pass between the fingers so as to touch all parts of the hand. The reason for cupping the hands is to provide for the washing of the whole hand with as small a quantity of poured water as possible. The metaphor Mark used to describe this position is ‘fist’; we, however, would use the metaphor of ‘cup’.Footnote 15
Reynolds has been influential, and rightly so. While it is not always noted in discussions of his solution, despite being one of his key points, Reynolds developed the argument that the use of ἐν had curtailed the instrumental dative by our period and so we should read Mark's πυγμῇ as a dative of respect.Footnote 16 While the closest biblical parallel to πυγμῇ in Exod 21.18 could potentially have provided a possible influence for reading an instrumental dative, reading πυγμῇ as a dative of respect is also in line with the practice of hand-washing we saw mentioned in m. Yad. 1.2. Collectively, this would suggest that in general terms Reynolds is along the right lines, at least in the sense of πυγμῇ referring to the position of the hand being held out in relation to the pouring of water.
However, there are also some clear problems. Reynolds’ argument would have been stronger if he were able to provide ancient evidence to back up his claim about ‘cupped hands’. Moreover, Reynolds did not provide any evidence that ‘fist’ was used in the metaphorical sense he described. Reynolds’ case is not helped by the standard uses of πυγμή. Πυγμή occurs twice in the LXX translating the only two occurrences of אגרוף (Exod 21.18; Isa 58.4). Exodus 21.18 even has the dative πυγμῇ we find in Mark 7.3, though more obviously the instrumental dative than Mark: ‘When individuals quarrel and one strikes the other with a stone or fist (באגרף/πυγμῇ)…’ The use here is obvious enough and likewise in Isa 58.4: ‘Look, you fast only to quarrel and to fight and to strike with a wicked fist (ולהכות באגרף רשע/καὶ τύπτετε πυγμαῖς ταπεινόν)’. These uses are entirely standard uses of πυγμή in the pugilistic sense we might expect and they have long been documented.Footnote 17
By contrast, there were other, more obvious, options open for Mark if he wanted to convey the meaning Reynolds wanted to ascribe to πυγμῇ in Mark 7.3. Perhaps the most obvious context would be m. Yad. 1.2, where something like ‘cupped hands’ is used in a related context: ‘Nor should a man pour [water] for his fellow with his cupped hands (בחפניו)’. חפןFootnote 18 would have been a ready-made Hebrew or Aramaic equivalent for Mark to have translated into Greek, used as it was in discussions of hand-washing to describe ‘cupped hands’. Furthermore, three different Greek words are used to translate חפן in the LXX which Mark could have used: δράξ (Eccles 4.6), κόλπος (30.4 [24.27]) and χείρ (Lev 16.12; Ezek 10.2, 7). Alternatively, another word used for ‘handful’, קמץ,Footnote 19 might have been another, better alternative. קמץ would also have been known from the priestly tradition, in the sense of taking a handful of choice flour (Lev 2.2; 5.12; 6.8; 11Q19 20.10; cf. 11Q20 4.21) and translated with δράγμα (Gen 41.47) and δράξ (Lev 2.2; 5.12; 6.8 [15]) in the LXX. But Mark, of course, does not use a known Greek equivalent to words such as חפן or קמץ; instead we have the direct Greek equivalent of אגרוף which in early Jewish literature, in addition to the two occurrences in biblical literature, is, like the Greek, simply the shape of a ‘fist’ with typical connotations of striking, violence and power,Footnote 20 including, again, the direct equivalent of the dative πυγμῇ, באגרוף (‘He is not to raise his hand and strike it with the fist’ [CD 11.6/4Q271 5i.3; cf. 4Q364 13 a-b.6]). In the face of all this, we have to conclude that it is too much of an exegetical stretch to make πυγμῇ a precise or metaphorical reference to cupped hands.
3. Minimal Water and Purification
What all this also implies is that there is a far stronger case for simply taking πυγμῇ literally, at least in terms of the general shape of the hand.Footnote 21 In addition to the linguistic arguments just outlined, there are further cultural reasons to support this literal reading. There is one key detail which Reynolds mentions almost incidentally but which is, I think, crucial to understanding πυγμῇ in Mark 7.3: the minimal amount of water required to make the hands pure.Footnote 22 The right amount of water is required for making sure an object, hands or body are genuinely made pure because if too little is used purification might fail. This is common enough in Mishnaic purity discussions. One particularly relevant example concerns the immersion of a ‘bed’ or ‘dining couch’ (מטה) in m. Miqw. 7.7:
[If] one immersed the bed therein, even though its legs sink down into the thick mud—it is clean, because the water touched them before [the mud did]. An immersion pool, the water of which is [too] shallow [to cover the body]—one presses down, even with bundles of wood, even with bundles of reeds, so that the [level of the] water may rise—and he goes down and immerses. An [unclean] needle which is located on the steps of the cavern—[if] one stirred the water to and fro—after a wave has broken over it, it is clean.
The immersion of a bed in the right amount of water is particularly relevant not only because it is the kind of object immersed according to Mark 7.4 but also because just enough water has to be used to make certain the object is pure, even to the extent of pressing down to make sure. This applies to immersion more generally and the context of the right amount of water in purity law is especially important because the key texts from m. Yadayim discuss the minimum amount of water required for washing hands (quarter-log = one-and-a-half eggs):
[To render hands clean] a quarter-log of water do they pour for hands, for one, also for two. A half-log [is to be used] for three or four. A log [is to be used] for five and for ten and for a hundred. R Yose says, ‘And on condition that for the last among them, there should not be less than a quarter-log’. (m. Yad. 1.1)
[If] one poured water for one hand with a single rinsing, his hand is clean. [If he poured water] for two hands with a single rinsing—R. Meir declares unclean unless he will pour a quarter-log [of water]. (m. Yad. 2.1)Footnote 23
The rabbinic texts are, obviously, all later but, as we will see, all the other practices in Mark 7.3–4 are precisely paralleled in later rabbinic literature and we are on relatively solid ground with the earliest rabbinic texts, at least in general terms rather than the specific dating of each rabbinic saying or description. This is because it would be remarkable if Mark had invented such traditions only for them to be picked up or reinvented by later rabbis. And to add to this argument of collective weight on the role of getting the amount of water correct we can finally turn to the more contemporaneous Dead Sea Scrolls where this was, perhaps unsurprisingly, crucial in one understanding of purity:
Concerning purification with water: no-one should bathe in water which is dirty or which is less than the amount which covers a man. No-one should purify a vessel in it. And every cavity in the rock in which there is not the amount which covers, if an impure person has touched it, he has defiled its water <like> the water of a vessel. (CD 10.10–13)
Whatever we are to make of water usage at Qumran, this text shows the underlying logic of using too little water (and hence why it was, in part, important to specify minimal use): in this case, too little only succeeds in defiling the water and nullifies the purification properties.Footnote 24
This context of the minimal amount of water for purification also helps explain the significance of washing to the ‘joint’/פרק in m. Yad. 2.3, a text which, as we have seen, seems to be part of the discussion of an economic use of water (m. Yad. 1.1; 2.1). If we take πυγμή in its typical and literal sense as ‘fist’, what we have is an ideal solution for dealing with the problem of covering as much of the hand with as minimal water as possible. In other words, what a clenched fist does is to allow the water to cover all around the fingers and seep through the gaps in a way that would not be so effective with (say) pouring water over a hand with the palm out as it would require turning the hand over and wasting more water.Footnote 25 Using the fist, then, provides the most effective use of water. While not a precise analogy, the concern for letting water flow between gaps in hand-washing is attested in a different way in m. Yadayim: They pour out [water on the hands of] four or five people side by side, or above one another, on condition that they [the hands] lie loosely (שׁירפו) so that water will flow among them’ (m. Yad. 2.3).
We can now conclude that πυγμῇ in Mark 7.3 is not quite as mysterious as often thought and the solution is (deceptively) simple. If we take πυγμῇ in its typical and literal sense, accept Reynolds’ emphasis on the grammatical reasons for accepting a dative of respect and combine these arguments with that concerning the only known practice of pouring water on hands for purity, and use the context of the minimal water requirements for washing hands, then we can say, with some confidence, that πυγμῇ in Mark 7.3 is a perfectly commonsensical turn of phrase for Mark to have used.Footnote 26
4. Πυγμῇ and Mark 7.3
In addition to solving a tricky interpretative issue in πυγμῇ, what might this seemingly minor detail tell us about the interpretation of Mark 7.1–23 more generally? For a start, πυγμῇ in Mark 7.3 probably also has the function of distinguishing the washing of the hands from immersion in a miqweh (cf. Lev 12.4; m. Hag. 2.5).Footnote 27 Πυγμῇ is also part of a longer editorial aside in Mark 7.3–4 where Mark describes other related practices, namely the immersion of cups, pots, bronze kettles and beds (βαπτισμοὺς ποτηρίων καὶ ξεστῶν καὶ χαλκίων [καὶ κλινῶν]), which are accurate and precise descriptions of known Jewish practices. The issues relating to the impurity and/or immersion of such objects, along with general utensils made from wood and metal, are all found in rabbinic literature as expansions and interpretations of Lev 11.32–33 and 15.12 (e.g. m. Miqw. 3.10; m. Kelim 2.1; 5.11; 11.1; 13.6; 14.1; 15.1; 16.1; 20.2; m. Hullin 1.6). The immersion of a κλίνη—which proved to be sufficiently confusing for us to accept that it is more likely to be original than a later scribal addition—was a known practice as it is explicitly discussed in m. Miqw. 7.7 and m. Kelim 19.1, both using the direct Hebrew equivalent of κλίνη (מטה).Footnote 28 That Mark has provided this detailed and accurate editorial aside is clear evidence of the kind of message Mark wishes to get across. It may be, as recent arguments have stressed, that this editorial aside is part of Mark's construction of a dichotomy between the commandments of God and ‘tradition’ (Mark 7.1–13) as the key context for understanding 7.15, 19 not as an attack on biblical food and purity laws but as an attack on ‘tradition’ or the expansion of biblical laws. According to this argument, Mark 7.19 is a declaration that all kosher foods are clean and not made unclean by avoidance of hand-washing.Footnote 29 This would mean that Matt 15.20 is an accurate interpretation of Mark 7.19: ‘These are what defile a person, but to eat with unwashed hands does not defile’.
Mark 7.1–5 is particularly important in this regard as these verses directly relate to issues of the transmission of impurity from unwashed hands to ordinary non-priestly food to the eater, probably to keep insides pure.Footnote 30 Ordinary food is most susceptible to second degree impurity, at a second remove from the scriptural source (e.g. m. T. Yom 4.1, 3; m. Sotah 5.2). The conventional rabbinic view holds that impure hands, deemed second degree impure, could not render ordinary food impure because something deemed second degree impure could not defile something else susceptible at most to second degree impurity (e.g. m. Yad. 3.1).Footnote 31 However, the role of liquids is crucial here.Footnote 32 The strong defiling function of liquid in the transmission of impurity is very specific in Lev. 11.29–38 but was expanded more generally in rabbinic literature—and elsewhere (e.g. 4Q284a; 4Q274 3; 4Q394 8.4, 5–7; CD 12.15–17)—so that something of second degree impurity such as unclean hands could pass on impurity to ordinary food (e.g. m. Parah 8.7; m. Zab. 5.12; m. Hullin 2.5; b. Hullin 33a). And the liquids which defile—namely dew, water, wine, oil, blood, milk and honey (m. Maksh. 6.4; cf. CD 12.15–17)—have obvious relevance for the meal table. Given that Mark 7.4 mentions bodily immersion, we should also note the issue of the tebul yom, i.e. someone who has immersed that day and is waiting for sunset to be deemed pure again (Lev 15). A tebul yom is deemed second degree impure (e.g. m. Zab. 5.3). However, a tebul yom cannot render liquid impure in contrast to unclean hands even though both are second degree impure. Yet at the same time it was possible for a tebul yom to have their hands second degree impure apart from the rest of the body and these defiled hands are able to defile apart from the rest of the body (e.g. m. T. Yom 2.2). And so washing πυγμῇ, and keeping vessels pure, would have been important even after bodily immersion. This process is precisely replicated in Mark's editorial aside in 7.3–4 adding further weight to the argument that Mark has an excellent knowledge of purity law and is offering a critique of hand-washing and not biblical food and purity laws.Footnote 33
5. Concluding Remarks
We have seen, then, that the most obvious and literal reading of πυγμῇ and an understanding of hand-washing in purity law, particularly the lower limit of water required for purification, means that we should understand πυγμῇ in Mark 7.3 as meaning ‘with the hand in the shape of a fist’ in order to be covered with as little water as possible. We also saw how this approach fits into Mark's detailed understanding of purity and hand-washing and plays its part in establishing a clear context to interpret Mark 7.15, 19 in light of the transmission of hand-washing. A side benefit of this sort of approach is that it highlights further the already problematic scholarly explanations of πυγμῇ as a scribal error. After all, only by ignoring the background in purity law could Skeat suggest that πυγμῇ was nonsensical and the result of Mark accidentally repeating ἐὰν μή, realising his error and then blotting out the letters but not to the extent of fully blotting out μη which then had to be dealt with by a professional scribe responsible for the archetype of the Gospel.Footnote 34 We can make similar criticisms of W. D. McHardy's even more speculative suggestion that πυγμῇ was originally a marginal note for Mark 7.10 as an indication of the source material in Exod 21.Footnote 35 These sorts of approaches to πυγμῇ making no real sense turn up regularly in the history of scholarship on Mark 7.3 but must be deemed completely inadequate. What the approach to Mark 7.3–4 advocated in this article also does is to contribute further to the undermining of the increasingly less popular idea that Mark is basically ignorant of Palestinian Jewish customs.Footnote 36 As we have seen in the case of πυγμῇ and all the other practices described in Mark 7.3–4, we are dealing with known Jewish practices of the time and Mark knows the details in depth. Whatever we are to make of Mark 7.1–23 in the future, it is going to be extremely difficult to claim that Mark did not know precise details of Jewish purity law. Purity law has proven to be one of the more difficult areas for NT scholars to understand but this cannot be an excuse to ignore its complexities when it occurs in the texts.