Theymade “ a mistake ” in Job , 4 , 11 ; why not also in Prov , 30 , 30 ? Implicitly , about limits in philology and the necessity of accepting them

The existence of the Greek term μυρμηκολέων in Job, 4, 11, in the biblical text of Orthodox tradition, on the steadfast line of the Septuagint, has seemed—time and again—bizarre, and has intrigued enough as to be approached as a textological problem in several articles and studies, some of them extended and well documented, during the last hundred years; it seems that we face an ordinary translation mistake: the Hellenised Hebrew translator of the Septuagint has missed the equivalent of the Heb. ל י ש (layish [lah’-yish]) ‘a lion’ from the original story of Job, and has produced, consequently, what appears to be a hapax legomenon in the sacred text, an odd and obscure term. Reopening the case, the present study argues in favour of a different reading of the word μυρμηκολέων, which precludes the translator’s presumed mistake. Moreover, it reminds the researcher of the necessity to question, in a lucid manner, the arguments that seem to support a certain conclusion.


An unexplained "mistake"
The existence of the Greek term μυρμηκολέων, οντος (ὁ) (cf. μύρμηξ, ηκος (ὁ) & λέων, οντος (ὁ); see bailly, s.v.; liddell-scott, s.v. μυρμηκ-; Engl. antlion; Rom. furnicoleu) in Job, 4, 11, in the biblical text of Orthodox tradition, on the steadfast line of the Septuagint (μυρμηκολέων ὤλετο παρὰ τὸ μὴ ἔχειν βοράν σκύμνοι δὲ λεόντων ἔλιπον ἀλλήλους, lxx; e.a.), has seemed-time and again-bizarre, and has intrigued enough as to be approached as a textological problem in several articles and studies, some of them extended and well documented, during the last hundred years (Druce, 1923; Kevan, 1992 1 ; Cardell, 2013;  Munteanu, 2016).The usual conclusion concerning the cause of its presence in a context that suggests as normal the option for lion-both from the point of view of the symmetry of the verse (cf. the second part: "...σκύμνοι δὲ λεόντων ἔλιπον ἀλλήλους", lxx; Engl."…and the stout lion's whelps are scattered abroad", kjv; Rom."...şi puii leoaicei se risipesc", b 2008; e.a.), and from the point of view of the immediate logic of the text-is that we face an ordinary translation mistake: the Hellenised Hebrew translator of the Septuagint has missed the equivalent of the Heb.ׁ ‫ִש‬ ‫֭י‬ ַ ‫ל‬ (layish [lah'-yish]) 'a lion ' (strong, s.v.), from the original story of Job (see strong,s.v. Job,4,11,interlinear: Hebrew), and has produced, consequently, what seems to be a hapax legomenon in the sacred text, an odd and obscure term.

Premises
There are numerous cases when, on one hand, having lost the contact with the world on the realities of which the original Hebrew discourse has been established, being unable to recuperate the knowledge about that world, and, on the other hand, feeling that he has a duty toward his contemporary reader, the translator "betrays" the source-text: either by a) making a plane mistake (he confuses the terms, attributes to them meanings that they do not have, and, consequently, translates them incorrectly into Greek); or by b) choosing what he things to be the most plausible equivalent of a Hebrew word in Greek, but knowing that he could be wrong; or by c) employing standard equivalents, since he usually practices a stereotypical translation, unconcerned about the possibility of being in error; or by d) taking the liberty to correct or to clarify the text, according to his own understanding and to the understanding of his public. 3.Accordingly, the evaluation of the Gr.μυρμηκολέων, οντος (ὁ) [Engl.antlion; Rom.furnicoleu] as a "lexical creation stemming from confusion" 4 (Munteanu, 2016, p. LXVI; our transl.), in Job 4, 11, is not necessarily strident, but in agreement with the normality of a text's transfer between two languages and, to some extent, two cultures that are wide apart.However, the generalization does not serve the truth, and the researcher cannot give the final judgement on an issue by dint of a single piece of evidence, especially when that piece of evidence is conjectural.
the impossibility of an actual confusion in Job, 4, 116 ; but there is no evidence for the existence of some linguistic, contextual and co-textual conditions that would favour the confusion: i.e. some situation of homonymy, paronymy, or the proximity of a perturbing term, etc.
Rather, we believe, one cannot refute the idea that, in Greek, Job, 4, 11 contains μυρμηκολέων, οντος (ὁ) as the proper equivalent for the Heb.ׁ ‫ִש‬ ‫֭י‬ ַ ‫ל‬ (layish [lah'-yish]), with a meaning close to the meaning of the original word, in the original source-text.. Should this be the case, one needs to re-examine the semantic evaluation of the Greek term: Is it possible it refers, in the spirit of the Hebrew text, to a creature whose distinctive physical features ([+ quadrupedal], [+ mammal], [+ predator]) are common to those of the unequivocal lion?
The Greek zoonymy does not seem to support this hypothesis.However, as is known, several Antique writers7 record the existence, in some place (India, Ethiopia, Mesopotamia…), of an animal in the description of which appear several elements that, as they pass from one text to another, along the centuries, converge toward the possibility that the consciousness of some readers living in the Ptolemaic epoch grasp the notion of an extra-linguistic reality that goes by the Greek name μυρμηκολέων, οντος (ὁ), from the lexical field of mammals, sharing with λέων, οντος (ὁ) a few substantial semes.
The hypothesis of a zoological confusion-expressed in the notes and the commentaries on some of the aforementioned books-alludes to a terrestrial animal of the Sciuridae or Herpestidae family (the marmot, the gopher, the mongoose...), whose area covers wide regions in the Middle East, India or Africa, and which produces, by digging, formations resembling the anthills, or which takes soil in underground galleries (see, for Herodotus: Barguet, 1964Barguet, , p. 1411;;Piatkowski, 1961, p. 489;for Aelian: Scholfield, 1959, p. 163).The issue has been approached time and again, and, in striving for the exact identity of the elusive creature, the scholars have followed numerous and various clues, from linguistic ones-observing the form and content of the Mahābhārata-, to ethnographic ones; the opinions were slightly different, but they converge to the verdict marmot (see a synthesis at Cardell, 2013).
However, we are less interested8 in the identity9 of the creature that happens to dig gold (see Herodotus, loc.cit.), as known by the human population of that mythical oriental Eldorado.Because, once entered the Greek discursive stream, and being used in a context that remains relatively stable 10 but upon which tells the consciousness (reasoning) of the receptor, the linguistic sign with which the reader/speaker has to operate is a Greek one, one that has (or receives) or not a certain meaning.The way in which Strabo, e.g., employs the termμύρμηξ, ηκος (ὁ), at the end of the 1 st c. bc, writing about exotic lands, points out a thorough judgement upon the reading that one should apply to this word, when it is found in a certain (con)text.It is probably improper to state that, by the beginning of the 1 st c., Gr. μύρμηξ, ηκος (ὁ) developed a new meaning for the ordinary speaker, forming a case of homonymy still unregistered by dictionaries.But it seems plausible that a certain type of speaker, culturally conditioned, be capable of recognizing in μύρμηξ, ηκος (ὁ) the Greek reflex of a foreign linguistic significant whose signified is 'lion' , and of activating the association μύρμηξ, ηκος (ὁ) -'lion' , in a context similar to that that generated it at some point.
Unfortunately, the claim does not escape the spectrum of hypotheses.Yet, we express it, for it argues for the competence of the translators of the Septuagint, not as lacking in intellectual ability and translating skills as we might think now, after more than two millennia of perpetually accumulated experience, knowledge, and lexicological bibliography. 11.

Other premises, and what follows from there on lexical level
The Hellenized Hebrew translator needed a "rare" word, like the one he found in the text he was translating, that would have indicated 'a sort of lion' , in a story whose dramatical action took place somewhere outside Israel, in a land how vague so ennobled due to the value of the moralizing story of Job, and whose descriptive and distinctive features, therefore, had to be preserved as such.Judging by the theme and the style of the Book of Job, the mythical land Uz (Job, 1, 1) is located somewhere in a vast area that includes Egypt, Mesopotamia, the south of Edom and the northern region of the Arabic Peninsula (Seow, 2013, p. 61, 314, 496, 702)-the area where, according to some of the ancient writers, lives the legendary 11 Studying thoroughly the matters concerning the translation of idiomatic expressions from Hebrew into English, Joosten (2010a) grasps the remarkable position of the Seventy in relation both to the source-language, and to the target-language: "on the whole, the Greek translators's grasp of the source language was excellent.Of course, the meaning of one or another Hebrew expression may indeed have been forgotten by the Hellenistic period.But on the whole, the translators understood the idiomatic expressions well enough: literal renderings are not to be attributed to a lack of understanding" (p.66; e.a.); and: "The way the translators dealt with idiomatic expressions also reveals something of their deeper motives.The translators brought great creativity to their project.Their objective, however, was not to create something new and unprecedented, but to preserve the old.To all appearances, the ultimate goal of the translators was to give to their readers as much as possible of what they found in the source text.Although the translational process sometimes demands that one should abandon either the wording of the source text or its global meaning, the Seventy were not at ease with this alternative.More often than not, they refused this basic dilemma and tried to compose in Greek an expression that paid tribute to both the wording and the sense.Although some of their renderings are open to criticism, because they follow neither the form nor the meaning of the source, they reflect much intelligence and a general preparedness to try out new formulas" (p.68; e.a.).
μύρμηξ.However, had the translator simply used μύρμηξ, ηκος (ὁ), he would have caused perplexity to the common reader, who would have thought spontaneously of the meaning 'ant'; or, a noun compound of two elements that were essential in that particular context, appeared as the perfect solution, both for the atmosphere of the original text, and for the new reader: *μυρμηκο-λέων (lit.ant-lion), the second element functioning as a clarifying synonym of the first element, the expected "reading" being: 'the ant in the sense of l i o n...' (or 'the ant which is actually a l i o n'), according to the logic of the whole verse.
In terms of form, the word belongs to a class of compound zoonyms quite common in the Hellenistic epoch and in later Greek (Bodson, 2005, p. 463): hippo-tigris, lit."horse-tiger"; hipp-elaphos, lit."horsestag"; kamelo-pardalis, lit."camel-leopard/panther"; stroutho-kamelos, lit."bird-camel", etc.But it doesn't necessary follow and it is not equally clear that it shares the same semantic substance with the aforementioned examples.In hippo-tigris,one recognizes a model in which "the names of two animals are placed side by side to identify a third one primarily seen by the ancient people as sharing some morphological and often behavioural traits with both of them" (Bodson, 2005, p. 463), namely: 'a horse with t i g e r-like stripes' (i.e., zebra), 'a camel with l e o p a r d-like patches' (i.e., giraffe), 'a bird that looks and runs like a c a m e l' (i.e., ostrich), etc.Such name inspiring descriptions occur in the presence of the animal that has to receive a name, and the namer knows well the aspect and behaviour of the three creatures involved in the process; moreover, numerous sources depicts the namee, and justify the chosen name in a particular case.12 .Or, concerning the μυρμηκο-λέων, if we were to accept this paradigm (the first element of the compound name indicates the genus, and the second element, the species, Bodson, 2005, p. 463) 13 , we would expect to find among the texts of the period one or more sources confirming the existence and describing the appearance of an actual creature that seems to be an ant (or, perhaps, an insect) with some characteristics of a lion ([+ robustness], [+ aggressiveness], [+ ferocity], [+ predatory technique]), a creature that has been wrongly perceived as being designated by the Heb.[lah'-yish].This doesn't happen but several centuries later, when, in the 6 th and 7 th c., and, more obviously, beginning with the 9 th c., scholars like Gregory the Great, Rabanus Maurus, Albertus Magnus and Thomas of Cantimpré (the two last-named, during the 13 th c.) describe a μυρμηκολέων, οντος (ὁ) / Lat.myrmēcŏlĕōn either as an ant larger than other ants, or as a particularly aggressive ant, that feeds on regular ants, or, finally, as a larva that feeds on the ants that slip to the bottom of its sandy trap. 14.
The interval of almost a millennium between the attestation (probably, creation, as well15 ) of the Gr. in the Septuagint, and its first employments with the meaning it has today, raises a question concerning the truthfulness if the idea that the Septuagint deals with the same μυρμηκολέων, οντος (ὁ) (Engl.antlion; Rom.furnicoleu/leul furnicilor) we find in nowadays entomology.
The lack of antique sources that should document the existence of a μυρμηκολέων, οντος (ὁ) as a 'real insect' (see supra, note 4) hinders also the idea that, at the beginning of the 2 nd millennium-after the late Antiquity and the Middle Ages in which various exegetes, the Physiologus and the Bestiary had created and popularized exclusively the image of a fabulous creature by the name of μυρμηκολέων-, scholars like Albertus Magnus did not achieved but a mere rediscovery of the original meaning of the word.
The phenomenon is known for kamelo-pardalis (Buquet, 2006(Buquet, , 2008)), stroutho-kamelos (Bodson, 2005, p. 467-472), etc., but the initial conditions specific to those cases are not repeated with μυρμηκολέων, οντος (ὁ)!As it happens with the "clarification" of the other compound names of animals, the Oriental patristic literature, the Physiologus and the Bestiary see μυρμηκολέων as an animal of double nature: with the body of an ant, and the head of a lion, that can feed neither on grains-because it is also a lion-, nor on meatbecause it is also an ant; consequently, and conveniently, the exegeses speculates it moralizingly.However, by dint of their word, one cannot conclude that what goes by the name μυρμηκολέων is, in itself, and along the centuries, "a fantastic animal from the Medieval bestiary, without a real existence" (Munteanu, 2016, p. LXIV; e.a., our trans.),and cannot find out either "what is, in fact" the antlion (idem, p. LXVII; e.a., our trans.),because: a) to a great extent, texts as such are formed circularly; b) it is highly possible that the described morphology of an animal might reflect a superficial etymological analysis/the superficial formal level of the word in question, many ears after its first occurrence in a text (in a case of folk etymology); c) the animal morphology itself finds justification in the importance one assumes it has in formulating and supporting certain spiritual precepts.
The last observation might suggest that μυρμηκολέων presents, in fact, a case similar with that of some compound animal names that occur in Greek literature anterior to Septuagint, that display the same structure, and which do not designate real living creatures: e.g.kunamuia lit.'dog/bitch-fly' , or kunalopex lit.'dog-fox' , but describe metaphorically a human type, i.e. "the annoying impudent" and "the impertinent"... Likewise, it is-one may say-possible that the inspired translator of the book of Job might have wanted to (re)create the image of a "hypocrite", of someone whose existence, because he/she is two things simultaneously, cannot be but denied16 by a lucid authority... (see the often made connection between Job,4,11 and Mt,5,37).And yet, at Homer and Aristophanes, the naming follows the need to characterize, and nothing suggests a different state of situation; while in the case of the Gr.μυρμηκολέων, the figurative interpretation, and, consequently, the characterization of the human nature follow the finding of the word, under the imperious need to give sense (a particular17 sense!) to the text.
On the other hand, it seems to us that there is an apposite similarity, although partial, between the case of μυρμηκολέων and the case of another compound name, tragelaphos lit.'goat-stag' , which, at some point, lost the meaning of 'fabulous animal/unnatural monster' , that it had had at Aristophanes, in the 5 th -6 th c. bc 18 , and found a place in the semantic field of the natural fauna (bailly, s.v. τραγέλαφος, ου, (ὁ) 2.: post.'sorte de gazelle ou d'antilope à barbe de bouc'), of the "natural monsters" like the kamelo-pardalis that we find at Diodorus Siculus, in the 1 st c. bc: αἱ δὲ καλούμεναι καμηλοπαρδάλεις τὴν μὲν μίξιν ἀμφοτέρων ἔχουσι τῶν ἐν τῇ προσηγορίᾳ περιειλημμένων ζῴων.τῷ μὲν γὰρ μεγέθει μικρότεραι τῶν καμήλων εἰσὶ καὶ βραχυτραχηλότεραι 19 , τὴν δὲ κεφαλὴν καὶ τὴν τῶν ὀμμάτων διάθεσιν παρδάλει παρεμφερεῖς διατετύπωνται: τὸ δὲ κατὰ τὴν ῥάχιν κύρτωμα παρεμφερὲς ἔχουσαι καμήλῳ, τῷ χρώματι καὶ τῇ τριχώσει παρδάλεσιν ἐοίκασιν: ὁμοίως δὲ καὶ τὴν οὐρὰν μακρὰν ἔχουσαι τὴν τοῦ θηρίου φύσιν ἀποτυποῦνται.2. γίνονται δὲ καὶ τραγέλαφοι καὶ βούβαλοι καὶ ἄλλα πλείω γένη δίμορφα ζῴων καὶ τὴν σύνθεσιν ἐκ τῶν πλεῖστον τὴν φύσιν κεχωρισμένων ἔχοντα, περὶ ὧν τὰ κατὰ μέρος μακρὸν ἂν εἴη γράφειν (Diodorus Siculus, 1888-1890, 2.51.1, 2;s.n., A.C.) ["The camelopards, as they are called, represent the mixing of the two animals which are included in the name given to it.For in size they are smaller than the camel and have shorter necks, but in the head and the arrangement of the eyes they are formed very much like a leopard; and although they have a hump on the back like the camel, yet with respect to colour and hair they are like leopards; likewise, in the possession of a long tail they imitate the nature of this wild beast. 2 There are also bred tragelaphoi (goat-stags) and bubali and many other varieties of animals which are of double form and combine in one body the natures of creatures most widely different, about all of which it would be a long task to write in detail" (Diodorus Siculus, 1933, 2.51.1, 2;s.n., A.C.)].

Conclusions
As soon as one analyses them in their contexts, terms like those previously mentioned (myrmex, hippos; hippotigris, etc.; kunalopex, etc; tragelaphos, etc.) support the idea that the semantic life they have or develop is indifferent to the mould they have once assumed; that a certain form does not always dictates the semantic substance of a word; that a form can produce a certain content, according to the needs of the moment and of the translator; that there isn't a unique formula for evaluating the existence and the purpose of a particular word in a text.Therefore, to postulate that the Greek term μυρμηκολέων, in Job, 4, 11, is to be related neither to some fabulous creature, nor to the antlion of our entomology book does not have a weaker chance of being true, than the opposite one; but, as we've tried to argue, on the contrary. 20