1. The Ambiguity of the Terms
The term ‘source’ as used in connection with a transmitted text of lesser or greater extent connotes, first—as in the everyday meaning of the term ‘source’—the aspect of the origin/provenance of the material in a general way. Specifically, this aspect of ‘source’ is concretized in many and different ways. A ‘source’ for a historian may be a single written document which is chosen as the object of research in order to shed light on the origin/provenance and contexts of past historical events and/or people. For Rendtorff, ‘the Old Testament is the single source from which we learn something about the course and context of [Israel's] history’ (1988: 1). In accordance with Rendtorff, it is possible to understand the whole Old Testament in practice as a ‘source’ in the restricted sense of a source for Israel's history, and in an unrestricted sense as a source for Jewish and Christian belief. According to A. Alt (1968: 183) it is necessary before historical questions are addressed to the biblical text to clear up the literary foreground for the historical field of work. As a part of the notion of source criticism, ‘source’ is understood differently again. This understanding has its roots in the rediscovery of ancient conceptions of history at the time of the Renaissance and humanism, and the resultant attention to ancient sources. The attempt was made to study these in their original languages. Biblical scholarship was also not unaffected by the modern movement ad fontes. The critical study of the Bible, led by Hebrew philology, noted duplications and contradictions in the Old Testament text, above all in that of the Pentateuch. In the search for the reasons for such duplications, etc., theories were developed. In connection with the term ‘source criticism’ the term ‘source’ came to designate written documents of greater or smaller length whose subsequent inclusion in, and addition to, the final text could have caused the duplications and contradictions that had been observed. By isolating such sources, for which terms such as documents or fragments were also employed, it was expected that light would be shed on the literary composition of the texts, and insights would be gained into the history, faith, and religion of Israel (cf. Kraus 1982: 25–8, 242–59).
1.2 Literary Criticism
The method by means of which the isolation of sources was attempted was first called ‘literary criticism’, and then simply ‘source criticism’. It was originally confined to the study of the Pentateuch. Its results culminated in the ‘newer documentary hypothesis’ formulated by Graf, Kuenen, and Wellhausen. Its further development (by Eissfeldt, Fohrer, and others) was called ‘the latest documentary hypothesis’. In fact, all these modifications to the documentary hypothesis, which have continued to the most recent times, indicate the weakness of literary criticism, in so far as it has been understood and practised as source criticism. For attempts to isolate literary elements have not led in many parts of the Pentateuch to the discovery of one or more sources. Additionally, there are often great differences in results despite the use of the same method (e.g. cf. in Gross 1974: 419–28 the variegated results of the literary analyses of the prose in Numbers 22–4). These divergences presumably have their origin in methodological criteria that are differently defined and employed. Nevertheless, literary criticism serves to establish the unity or lack of unity of a text (cf. also Gross 1974: 143–7). Literary criticism thus constitutes the first methodological step on the path to seeking the origin and the provenance of a text. Further steps (including form criticism and redaction criticism) must follow later, in order to place a stratum or source in its proper place within a larger textual whole where possible (cf. Richter 1971: 50). Here, the term ‘literary criticism’ is used rather than ‘source criticism’. It goes without saying that with this narrowing of the term no option for or against the documentary or source hypothesis, which to this day is disputed or challenged, is implied (see Zenger 2001: 113–18 for the present state of the discussion).
2. Subject, Definition, Aim
Prior to the description of criteria, their use, and the appraisal of their demonstrative ability, it is necessary to consider the subject, definition, and aim of literary criticism. The order in which they are considered follows from the context in which they are grounded.
The subject, in the broadest sense, is the text that is transmitted in writing in the Hebrew/Aramaic Bible, to the extent that its literary unity is not immediately self-evident. Since this evidence of unity is the exception rather than the rule, every Old Testament text which lacks this evidence must be investigated in the context of a scientific literary-critical interpretation. In the narrower sense, every literary complex can be seen as the subject of literary criticism to the extent that this single text or text-complex lacks evidence of literary unity (cf. Richter 1971: 49; Barth and Steck 1987: 13; Fohrer et al. 1993: 45–6; Kaiser 1975: 23; Schmidt 1991: 211; Gross 1995: 648–9; Utzschneider and Nitsche 2001: 233–5). An opposite view is found in Dohmen, who, referring back to B. Jacob (1928 and 1930) is sceptical about the process of isolating literary sources, at any rate of that which is understood as source criticism. When he says that ‘the value of such analysis’ must be measured by the ‘plausibility of the (methodologically) necessary synthesis’, it is necessary to take notice. Is not this synthesis also agreed upon by all the writers just mentioned? The breadth with which Dohmen cites Jacob's opinion (2001: 89–92) evokes the suspicion that by this synthesis he means the ‘final form’ of the text, since he shares Jacob's scepticism about discovering any possible literary growth of the text.
Literary criticism is understood here as the method by which the literary integrity of a text is examined by means of a number of criteria that are relevant to the subject (i.e. a written text). At each stage of its procedure it is a synchronic method. Both the use of the criteria and the establishing of the lack of unity of a text in each case, as well as the resulting determination of the relation of the parts of the text to each other (relative diachrony) have to take place strictly at the level of the text under investigation, i.e. in the domain of synchrony. Literary criticism understood as synchronic is thus distinct from conventional source criticism, which improperly mixed synchrony and diachrony and was connected with the search for sources always according to their age and their subsequent redactional incorporation.
The aim of literary criticism comes from the definition of this term within its methodology. It lies in answering the question of whether the text that is being analysed is a literary unity or whether there are indications that it is composite. If the latter is the case, the aim of the literary-critical investigation is expanded to embrace the relative diachrony of the various elements of the text. Synchronically, an answer to the question is sought as to which textual element (or elements) is (or are) the presupposition for other textual elements. At this stage of the analysis the decision is confined to determining the original unit (Ausgangseinheit). Richter calls this ‘the older unity’ (1971: 70). Although he understands ‘older’ in a relational sense, the term ‘original unity’ will be used instead of ‘older’ in order to avoid misunderstanding.
Richter (1971: 70–2) gives (as examples) three cases which make possible the reaching of a decision. (1) Larger unit versus a single sentence or sentences: because the latter can hardly have existed as isolated cases, they are allocated to a later stage in the relative diachrony than the larger unit. The decision can be firmed up in so far as these single sentences indicate that they are additions (through such things as repetitions) to the greater unit. In addition, there are also more extensive insertions, which presuppose the original unity (compare the examples and their evaluation at Richter 1971: 70–1). (2) The original unit has been enlarged either at its beginning and/or at its end: these parts of the text then prove to be a frame for the original unit. These framing parts thus presuppose the unit which they frame (cf. the examples from Judges, Richter 1971: 71). (3) Two small units have been worked into each other: in this case each must have existed previously as a separate entity. Determining their relative diachrony is then not possible. This, however, could still be achieved if ‘one was composed in the view of the other’ (Richter 1971: 71).
3. The Criteria
Richter 1971 set the standards for the criteria of the method, its certainty, and its utility. Otto Eissfeldt, one of the old masters of classical literary criticism, according to Knierim (1985: 129), replied to G. von Rad's request in the early 1950s for a guide to literary criticism for students with a reference to his own introductory handbook. Richter (1971: 51 nn. 5–7) also referred to the ‘arguments’ of Eissfeldt and other scholars occasionally mentioning criteria. Richter can be credited with establishing and grounding the ranking order of the criteria, their certainty, and their utility. They are (1) duplications and repetitions, (2) contradictions (3) further observations, and (4) parallels to other parts of the text (for copious examples cf. Richter 1971: 51–62). The first criterion has the greatest weight. This weight becomes less with a rise in the number of criteria. ‘Certainty in a decision is scarcely possible with a single one of these indicators’ (Richter 1971: 62). It is not only the cumulation of criteria that increases the certainty. A decision is most certain when all criteria are present, although this is seldom enough. Arneth (2002: 389) sees the ‘common indices for the presence of literary growth taken from the data’. From those that are indicated, only doublets, syntactic breaks, and contradictions are included here.
‘Differences in terminology’, ‘differences in style’, ‘peculiarities of speech’, ‘elements typical of a genre’, occasion further analytical steps: for example, form criticism. As in the case of form criticism, Richter's methodology has produced ‘disciples’. Fohrer and others (1993: 45–58) do not deviate essentially from Richter. Also, Barth and Steck (1987: 34–8) agree to a considerable extent with Richter concerning the aim (examining the literary integrity) and the necessity of literary criticism. Also, even though in a modified way, they agree about the limiting of the criteria. As opposed to Richter, Barth and Steck deal with the question of the greater literary contexts and strata as well as their identification and historical ordering on the level of literary criticism, although they admit that further methodological steps are eventually necessary. Dohmen (2001) interprets this requirement quite liberally. He deals with the list of well-known repetitions and contradictions in the story of the Flood by means of arguments drawn from semantics, narrative logic, and narrative technique, and ‘on the basis of the rich treasury of ancient oriental stories of the flood’ (p. 101). Gen. 6: 9 (‘these are the generations of Noah …’), ‘which inserts a midrashic type of addition’ (p. 101), allegedly interrupts the context of Gen. 6: 5–8. These exclusively noetic arguments possess neither literary nor redaction-critical weight. Regarding the appropriateness, utility, and certainty of the criteria, and with them the method in general, note must be taken of the very important monograph of G. Vanoni, ‘Literary Criticism and Grammar’. In the course of his investigation of the repetition and contradictions in 1 Kings 11–12, Vanoni draws attention not only to the necessity of literary criticism as the very first step of analysis (1984: 18–21). He is able to demonstrate the priority of textual criticism initially within a single strand of the transmitted text (e.g. only in the strand of MT) in its positive consequence for literary criticism. By the textual criticism of several strands of tradition (e.g. the MT strand and the LXX strand at the same time), relevant literary critical indices may often, he says, be clarified in the course of textual criticism. A text thus ‘cleared up’ might possibly appear as a unit, while the MT in itself must be regarded as composite (cf. 1984: 269). ‘An examination that integrates the textual and linguistic features as much as possible has been valuable for literary criticism’ (1984: 169). The innovative function which literary criticism acquires through grammar, and which Vanoni convincingly expounds, can be explained again through the impetus which came from Richter's methodology in 1971.