1 Interpretive Quandaries of Mark 2.21

Following on from his appeal to the incongruity of fasting with wedding celebrations, Mark’s Jesus shifts abruptly to two material examples of incongruity.

In the immediate context and at some distance, the gist of these analogies appears tolerably clear: the newness signified by Jesus and his ministry is out of step with the oldness signified by the fasting of John’s disciples and the Pharisees.Footnote ⁸ And yet, on this reading ‘the conclusion of the two parables are certainly different: new wine goes into fresh skins, but an old garment presumably needs to be patched with a piece of old material.’Footnote ⁹ We can add to this that a new garment presumably would not need any patch at all, making the presenting issue of relative fit irrelevant. Moreover, the phrases left untranslated above present several quandaries. I will take them in order.

The first problem is determining how to understand the phrase ἐπίβλημα ῥάκους ἀγνάφου. The opening word in this phrase is a general term for a covering which is often used more specifically in relation to textiles.Footnote ¹⁰ In the present context, as the object of the verb ἐπιράπτειν (to sew) and with its textile material elucidated by the subsequent genitives, reading ἐπίβλημα as a patch to cover an implied tear in the old cloak is generally accepted. The second term, ῥάκος, broadly denotes a strip of cloth though outside the context of medical treatises it tends to carry the connotation of ‘rag’ or ‘tatters’, from Homer’s depiction of Odysseus’ ragged disguise to Epictetus and beyond.Footnote ¹¹ Josephus, for example, uses the term to refer to the strip of cloth cut by David from Saul’s robe (Ant. 6.285–9). It is worth noting, moreover, that outside these gospel texts and their later commentators, ῥάκος never modifies ἐπίβλημα.Footnote ¹² In light of this, however, it is not clear in what way Mark can refer to this ‘rag’ as something ‘new’ (καινός) in contrast to what is old (the cloak).Footnote ¹³ Even in medical contexts, a (hopefully) clean – even (possibly) fresh – ῥάκος is not referred to as ‘new’.Footnote ¹⁴ This pattern of use belies the presentation of information in the BDAG entry. There, uses in medical contexts are not differentiated from others, while the passage from Josephus noted above is combined with a reference to Jer 45.11 (where the ‘rags’ are specified as ‘old’) as justification for interpreting Mark 2.21 as ‘a patch made of a piece of new cloth’.Footnote ¹⁵ ‘Piece’ is an unwarranted gloss for ῥάκος, and it obscures the question facing readers of this verse: why would Mark refer to this ῥάκος as ‘new’? This quandary is only deepened when we turn to the third term, ἀγνάφος, which normally applies to an item of clothing which has not been to the fuller, and therefore is implied to be new, a point to which I will return shortly.Footnote ¹⁶ What, then, is this ‘new’ rag of unfulled cloth and where does it come from?

The second quandary relates to the phrase αἴρει τὸ πλήρωμα ἀπ’ αὐτοῦ, in which both the subject of the verb and the meaning of the term πλήρωμα are unclear. Taking the second issue first, given its connotations of completeness, it is unusual for πλήρωμα to refer to a rag (ῥάκος), and yet the apparently parallel movement of the πλήρωμα away from some object (ἀπ᾽ αὐτοῦ) and the phrase τὸ καινὸν τοῦ παλαιοῦ suggest together that the πλήρωμα is associated with the new object, namely the ἐπίβλημα ῥάκους ἀγνάφου. Generally, the lexical issue is resolved by taking the term to indicate that which fills a given space (rather than completeness itself), though this remains problematic insofar as the evidence for such a definition is not clear. Both LSJ and BDAG point to Mark 2.21 for this definition though the secondary example in LSJ does not support such a ‘filler’ interpretation.Footnote ¹⁷ The entry in BDAG is even more problematic. Under the gloss ‘that which makes someth. full/complete, supplement, complement’, it opens with a citation from Appian (Mithr. 47.185), in which πλήρωμα denotes a full sum of money, and follows this with the comment ‘lit. of the patch on a garment Mt 9:16; Mk 2:21’. There is a manifest mismatch between the first example and the ostensibly ‘literal’ application of a supposed common meaning in the synoptic passages: in the former, the term speaks to completeness as such (the full, final sum of money) rather than something supplemental that makes an item complete. In support of its interpretation, moreover, BDAG cites F. C. Synge’s 1844 article on the passage from Expository Times, though in that piece, Synge argues specifically against identifying πλήρωμα with the patch; for him, it is more plausibly taken as the whole of the new garment.Footnote ¹⁸ In light of this lack of πλήρωμα-as-filler evidence, perhaps a better approach would be to take πλήρωμα as designating the entirety of the patch that lifts away from the old garment (rather than the patch itself tearing), though the identity of the ‘fullness’ remains unclear.

Further, while the majority of commentators and translations take τὸ πλήρωμα to be the subject of the verb αἴρειν, this leaves the formally active and normally transitive αἴρει without an obvious object.Footnote ¹⁹ On the other hand, no other clear candidate presents itself as an alternative subject for the verb, though scholars have suggested either the person implied in the opening οὐδεὶς, a reading anticipated already by some medieval interpreters or the patch (ἐπίβλημα) itself.Footnote ²⁰ If either of these arguments hold, however, neither helps to make sense of the clause or the scenario more generally: either the person who sews on the patch now takes it away, for reasons left unexplained, or the patch itself takes the full (old or new) garment away from something else (but what?).Footnote ²¹ Alternatively, the more common view that the new patch (as a whole) lifts itself from the old garment presents an intelligible scenario but forces one to read both πλήρωμα and αἴρειν in unusual ways. However one looks at it, the Markan version of this analogy is difficult to parse.

Finally, the third quandary relates to the tear (σχίσμα): why is it worse after the new patch pulls (itself) away from the old garment? Is it because the patch was sewn to a larger area that is also now torn? Or is it because the patch is itself now lost as well? Despite the evident concern for preserving what is new, does this ‘worse tear’ suggest a care for the old garment as well? What is it about being unfulled that makes the rag so unsuitable as a patch for the old cloak?

As we shall see in due course, Luke’s version clarifies each of these quandaries in its own way. Before turning to Luke 5, however, it will be helpful to locate the Markan language within the context of ancient fulling to better understand the framework within which Luke’s rewriting works.

2 Ancient Fulling: Processes and Vocabulary

While there was long scholarly misunderstanding of the process, working out of a familiarity with the medieval practice, ancient fulling was not aimed at felting woollen products per se, nor is there evidence that fullers functioned as part of the textile production process itself.Footnote ²² Rather, in the ancient Greek and Roman world, fulling (γνάφειν, γνάπτειν, κνάφειν, ars fullonia) was the process for preparing a newly sewn garment for wear and for cleaning/refurbishing a worn garment, especially those made from wool, but also occasionally those made from silk or linen.Footnote ²³

At the risk of some generalisation, the fulling process proceeded as follows.Footnote ²⁴ The work began with the soaping process, the aim of which was to work detergents (such as urine and fuller’s earth) through the fabric.Footnote ²⁵ This stage involved placing the garment in a tub with some detergent and then working it roughly with feet and hands. In the Hippocratic De diaeta, this part of the process is described as follows: ‘And the fullers accomplish (their work) in this way: they stomp, they beat, they pull.’Footnote ²⁶ This stage is also famously depicted in a frieze from house VI 8, 20–21.2 at Pompeii, in which four workers are standing in tubs, with one appearing to stamp on a garment while the others appear to be wringing theirs out. Once the dirt and grease had been suitably removed, the clothes were thoroughly rinsed to remove the detergents before they were dried, either over poles erected in the fullery walls or over a moveable wooden frame. (As is often the case with wooden material from antiquity, none of these have survived, but there are graphic depictions of fulling in which both drying options appear to be at work.Footnote ²⁷ ) The use of bats or clubs for felting the wool, as was common in medieval practice, is not evident in antiquity.Footnote ²⁸ Nevertheless, while not primarily the aim, some shrinking and felting of the woollen fabrics would necessarily have happened in the laundering process.Footnote ²⁹

As well-known as these laundering activities are, however, they were less applicable to new garments than to those already used and ‘must be seen as basic, preparatory stages’ for the main part of the fulling process: polishing the garment.Footnote ³⁰ This involved in the first instance carding the garment, that is, scraping the cloth to raise the nap (that ‘layer of projecting threads or fibres on the surface of a woollen or other textile fabric’) and then shearing off the raised nap to create a smoother, more uniform finish.Footnote ³¹ According to a common ancient etymology, the term γνάφος was derived from the verb κνάω, meaning ‘to scratch’, because this was the activity particularly associated with fullers.Footnote ³² Shearing was apparently necessary for new garments as well because the nap on an unfulled cloak could be extensive.Footnote ³³ After shearing the nap, some garments may have been placed in a press, though little about this part of the process is known or sure.Footnote ³⁴ At other times, or for particular garments, various other finishing processes appear to have been used, including burning sulphur under the clothing (either to bleach white garments or for some other softening effect).Footnote ³⁵ It is unclear, however, whether the sulphuring activities preceded or followed the polishing process.

In Greek texts, the most common item listed for fulling is the ἱμάτιον, and it appears to be the case that some other types of clothing were not fulled, either for financial, practical or aesthetic purposes.Footnote ³⁶ In other words, and strictly speaking, it was not a necessary process and some garments may never have been fulled. There was a social status element to having finely fulled clothing, as opposed to only dressing in unfulled (or even dirty) garments.Footnote ³⁷ Julius Pollux’s discussion of the dress for different stage characters even suggests that the state of one’s clothes – fulled or unfulled – carried a kind of semiotic weight, insofar as certain professions on stage are signified by varieties of fulled or unfulled garments.Footnote ³⁸ That fulling was practised on new garments despite their usability off the rack (so to speak), also indicates a certain aesthetic and therefore social element to fulled clothing.

In papyrological lists of clothing and household items, ranging from the 3rd century bce to the 2nd century ce, various garments are often (though not always) identified as fulled (γνάφος or ἀπὸ γναφείου) or unfulled (ἄγναφος). P. Merton 2.71 (c. 163 ce), for example, lists a number of items and their prices including a white, spartan style robe (στολὴ λευκὴ λακονησήμου) and white gown (σύνθεσις λευκή) which are both noted as being ‘from the fullers’ (ἀπὸ γναφείου), while the white shifts (κιτῶνες λευκοὶ) and the blanket and white cloaks (λῶδιξ καὶ ἱμάτια λευκά) are unfulled (ἄγναφοι).Footnote ³⁹ The fact that cloaks (ἱμάτια) were commonly fulled, however, suggests that the designation ἄγναφος here indicates that they would not remain unfulled, only that they were currently unfulled, which is to say, new. In the later Prices Edict by Diocletian (c. 301 ce), prices for the production and treatment of certain garments are specified.Footnote ⁴⁰ While the prices need not detain us here, the terminology in the Edict is in keeping with earlier papyrological material, which helps to establish the fact that the processes and terminologies did not shift dramatically between the first and third (or early fourth) centuries. In particular, ἱμάτια and other larger, external garments still appear to have been commonly fulled, even when new; every garment being fulled except one in Edict 22 is identified as ‘new’ (rudis/καινός).Footnote ⁴¹

In turning back to Mark 2 and leading into the discussion of Luke 5, there are a few important take-away points from this discussion of fulling. First, the fulling process was not an essential part of woollen textile production, as in later periods, but rather was concerned with either cleaning and polishing old clothes or simply polishing newly sewn garments. In other words, Mark’s language of an ἐπίβλημα ῥάκους ἀγνάφου that is καίνος raises the question: Why is a piece of cloth identified as being ‘unfulled’ and new unless it came initially from an already sewn garment? Second, and following on from that question, a garment that is ‘unfulled’ (ἄγναφος) has never been cleaned or finished (at least, not professionally), and while some garments may never have been fulled, newly sewn garments were naturally a large portion of the ‘unfulled garment’ category. The language of ‘unfulled’ even at times appears to be used for signifying that a garment is new. Third, the fulling process, during which the garment is washed in preparation for the principal tasks of carding and shearing, will have inevitably led to some shrinking (at least for woollen garments if not for the occasional linen or silk clothing sent to the fullers). With those points in view, let us turn to Luke 5.36 to see how it offers interpretive solutions to Mark’s quandaries.

3 Interpretive Solutions in Luke 5.36

Luke’s version of the cloth and wine parables are as follows.

We noted above that identifying Mark’s ἐπίβλημα ῥάκους ἀγνάφου was made difficult not only because of an under-determined quality of some terms, but more specifically because of the fact that the patch was simultaneously identified as a rag, unfulled and new. If, as we have seen, the fulling process was concerned with garments already sewn, then material designated as ‘unfulled’ will normally have been a completed garment. Indeed, as the evidence presented above shows, under certain conditions, the term ἄγναφος could itself indicate that a garment is new. When this use of ἄγναφος is read in conjunction with Mark’s designation of the patch as ‘new’ (καίνος), Luke’s account becomes legible within its predecessor’s. On this reading, Mark’s ‘patch (made from) a rag from an unfulled garment’ has been rewritten in Luke as a ‘patch torn from a new cloak’. This new telling, then, is neither a ‘transformation’ nor ‘complete reinterpretation’ of the Markan statement (except to the extent that every interpretation offers a kind of ‘complete’ reinterpretation), but it does clarify an otherwise highly allusive appeal to fulling practices and terminology. Whether or not Luke considered this new version to be clearer than Mark or, if so, whether he considered this a necessary clarification as opposed to a stylistic change, is difficult to determine. If the argument above about the ancient context of Mark’s fulling language is correct, though, then there is no reason to suppose a particular opacity for Mark’s terminology in the phrase ἐπίβλημα ῥάκους ἀγνάφου. It remains the case, though, that certain syntactical quandaries would have remained even for ancient readers.

When we consider the second and third quandaries outlined above, relating to understanding the phrase αἴρει τὸ πλήρωμα ἀπ’ αὐτοῦ and the excessive quality of the new tear in Mark 2.21, we can see how these are rewritten by Luke in such a way as to remove the apparent oddities in the Markan account while keeping in line with contemporary fulling practices. In the first place, Luke removes Mark’s apparently unusual use of πλήρωμα and αἴρειν and replaces them with the phrase τὸ καινὸν σχίσει, in which the subject of the verb is the person implied in the earlier ‘no one’ (οὐδείς) and the formally active σχίζειν has a clear object, τὸ καινόν.Footnote ⁴² Further, the nature of the poor fit between the patch and the old cloak is clarified. While Mark laconically states the problem with the phrase ‘the old from the new and the tear becomes worse’, Luke explains that the ἐπίβλημα torn from the new garment ‘does not fit with the old’ and also tacitly clarifies why the tear is worse. In Luke’s version, there are now two clearly delineated items harmed by this rash action: the new is ruined and the old remains unfixed. It is important to note, though, that neither the new nor the old is explicitly preferred here.Footnote ⁴³ It goes without saying that one does not (normally) purchase a new garment and tear it to rags to repair old cloaks, but it does not follow from this that old cloaks are not valuable. Indeed, the fact that the owner in the parable wants to repair his old cloak points toward its continued usefulness. The old garment is not itself a problem; it only becomes one when the owner lacks the wisdom to recognise that a new (unfulled) garment would be ill-suited for the repair.

In short, Luke’s version represents a reading of Mark’s ‘unfulled patch’ in light of ancient fulling practices, carried out on sewn garments rather than raw cloth. The vocabulary and syntactical changes noted here work together to draw out the implications of this recognisable social context even more clearly in order to emphasise the foolishness of the person ignorant enough to attempt such a repair.Footnote ⁴⁴