Cognitive Linguistics And The Tripartite Division Of The Law

By Peter D. Myers

[Peter D. Myers is a Church of England ordinand currently studying for a Masters degree at Oak Hill Theological College, London.]

The doctrine of the tripartite division of the Mosaic Law into the categories moral, civil, and ceremonial has been largely rejected by evangelicals today. Despite the doctrine’s early attestation in Justin and Ptolemy;[1] support from theologians such as Augustine, Cranmer, and Calvin;[2] and almost universal attestation in Protestant confessions of the Reformation era,[3] the doctrine has joined the ever growing number of passé teachings in the Reformed confessions which good evangelical ministers feel comfortable ignoring. This is not to say there has been a focused reconsideration of the issue, but rather that for various reasons, some of which I shall outline below, side-lining and rejecting the doctrine has become the evangelical default.

It is my belief that evangelicals (particularly Reformed evangelicals) should seriously reconsider their views on this matter, and I shall seek to demonstrate that there is a sound methodological basis for doing so through the application of cognitive linguistics to the writings of the Reformed divines and to Scripture. A justification for employing cognitive linguistics will be given in section 1, but I shall give two reasons here why the doctrine is worth re-examining at all.

First, the doctrine has been rejected largely as a by-product of responding to the New Perspective on Paul (NPP), and without understanding it as expressed in its best form; nor has any attempt been made to advance the doctrine in light of modern developments in theology and linguistics. The idea that when Paul rejected “works of the Law” in Galatians he was speaking specifically about Jewish identity markers was seen by many as a threat to the doctrine of sola fide.[4] The response of the neo-Lutheran school which emerged was to undercut this claim by asserting that Paul viewed the Law as a unity. Despite the fact that the influential treatments of the doctrine by Carson and Westerholm are subtle and nuanced (neither absolutely denies a distinction between moral and ceremonial), they take place in volumes written in response to the NPP, do not address the lexical basis for the distinction, are almost exclusively centered on the NT, and are (anecdotally) appropriated by some to be saying that the distinction is completely invalid.[5] This, indeed, is Moo’s position, also developed in response to the NPP, which became particularly influential through his essay in Five Views on Law and Gospel.[6] The debate surrounding the implications of such work on biblical theology, and particularly the use of the Law in the Christian life, appears to have decoupled from the NPP debate somewhere around the late 80s to early 90s.[7]

Secondly, dissolving the boundary between any two categories of the tripartite division has had tangible theological and pastoral implications on the church in recent history. Conflating the moral and ceremonial renders the Law functionally useless as a revelation of God’s ethical nature. What is ethical in the Law can only be identified retrospectively, and preaching from the Law is effectively reduced to expounding the most relevant NT passage—else how can one be sure that a given application really does continue under the New Covenant?[8] Several colleagues of mine have observed that seminary students and pastors influenced by neo-Lutheranism often explicitly vocalize their suspicion of the OT as a source of ethics and doctrine. Moo, Wells, and Zaspel have published similar statements.[9] Unless one totally rejects the Law as a moral guide for the Christian life (see the discussion on Sanlon below), its moral and ceremonial aspects must be distinguished. Conflating the moral and civil aspects of the Law is straightforwardly seen in the Theonomy movement, where Gentile governments are treated as virtual replications of the Israelite theocracy.[10] Conflating the civil and ceremonial aspects of the Law may be the error of some eccentric aspects of the Federal Vision movement. While there is no space to develop this suggestion here, I encourage the reader to reflect on the implications of the FV as a renewal of Christendom theology.

There is a need for modesty on two fronts, both because the scope of the problem being confronted is huge, as current winds of opinion drive strongly against the doctrine, and because the solution I shall present is ambitious—necessarily so given that criticisms of the tripartite division are often methodological and foundational in nature. My aim is to demonstrate that the doctrine in its classically Reformed expression can be set on firm methodological grounds. This will be done in sections 1 and 2. Section 1 introduces cognitive linguistics and five key principles pertinent to an appraisal of the tripartite division. Section 2 uses this linguistic perspective as a lens with which to examine afresh the classically Reformed descriptions of the doctrine. Section 3 demonstrates the application of this methodology to establish the beginnings of a scriptural foundation for the tripartite division from the OT. Naturally a definitive scriptural case would require an entire monograph, yet this foundation will suffice at least to undermine the prima facie assumption that the tripartite division is untenable.

To be more specific about what we expect to achieve, let us take up one of the clearest challenges to the tripartite division, which neatly draws together the most common arguments, namely, Peter Sanlon’s popular essay “The Curse of the Law.” Sanlon recognizes that many Christians “use the Ten Commandments or Sabbath laws to shape their life and help them grow in godliness.” But he asserts that “to do so is a denial of the New Covenant gospel.”[11] While this statement is extreme, I respect Sanlon for his rigorous consistency and passion to protect the reality of grace in Christians’ lives.[12] When addressing the tripartite division, he describes it as “deeply flawed” and helpfully sets out four such flaws, which to a greater or lesser degree reflect popular evangelical opinion:

1. The Law is viewed by Scripture as “a single undividable entity.”
2. The “question of what exactly goes into each division is impossible to settle.”
3. “Obedience to Law is the opposite of faith in Jesus and the work of the Spirit.”
4. It “is difficult to see how Jesus could fulfil only [the civil and ceremonial] parts of the Law and not a moral part.”[13]

By the end of this study, I hope to have demonstrated that point 2 does not stand when a more sophisticated understanding of linguistics is adopted. Regarding point 4, we shall not address biblical theology directly, but as a more accurate picture of the Reformed doctrine is uncovered, I hope that its nuances and complexities will be seen to require a more sophisticated refutation.[14] Section 3 will be particularly relevant to point 1, which will be addressed, but with the caveats already mentioned. The third point would require a completely different study, one focused on the NT rather than the OT.[15]

For many neo-Lutherans the NT is the near-exclusive source of pertinent data for this issue (particularly Galatians); therefore, some may question the relevance of an argument built on the OT, such as the one offered here. In reality, the overturning of current opinion against the tripartite division would require a range of studies that taken together address all four of Sanlon’s objections, and as such mention should be made of Philip Ross’s recent and very fine defense of the tripartite division on the biblical theological front.[16] Nevertheless, the foundational methodological issues addressed by the present study have significant bearing upon several factors, including the presuppositions taken when interpreting relevant NT data. Therefore, it is to those methodological issues we now turn.

I. Introducing Key Concepts From Cognitive Semantics

It has been the lament of Barr and Porter that biblical linguistics has failed to track with developments in linguistic science.[17] Furthermore, biblical scholars do not always state the theoretical linguistic basis assumed by their work (and are often even unaware of such assumptions) which can make it hard to track the relationship between the two fields.

For this reason, in order to understand properly why cognitive linguistics has been chosen as the tool to address the problem at hand, an introduction to the major developments in Hebrew semantics and linguistic science in the last century is necessary. We shall choose as our theme for this narrative the word-concept problem: how do words relate to things in the real world, or at least our perception of them?[18] I hope to demonstrate that the answer of cognitive linguistics establishes a sound methodological basis for examining the tripartite division, and then introduce five key cognitive linguistic concepts through which we shall examine the Reformed descriptions of the doctrine.

1. A Narratival Overview Of Hebrew Semantics And Linguistic Science

A seminal point of connection between biblical studies and linguistic science came in 1961 when James Barr drew attention to the “uncriticized philosophy of language” employed by biblical philologists of his time.[19] Barr objected to the “unsystematic and haphazard nature” of the methodology employed by those connecting theological concepts with biblical words.[20] He specifically illustrated his concerns using Kittel’s Theological Dictionary of the New Testament, which assumed that word and concept are identical.

Barr’s critique of Kittel is a textbook example of structuralist criticisms of the historical-philological school.[21] Historical-philologists tended to describe lexical semantics diachronically (hence their intense focus on etymology), and drew direct lines between the way a language functioned and the way its speakers thought.[22] This marriage of diachronics with psychology and culture is illustrated brilliantly by the preface to Koehler-Baumgartner which sees word and concept united by the historical transmission of meaning through “a tradition of understanding.”[23] Brown-Driver-Briggs (BDB) is also a classic testament to the historical-philological school, arranging lexemes etymologically by root.[24] The TDOT, the OT equivalent to Kittel, is still unfinished in English and is therefore the last bastion of scholarly historical-philology.[25] Despite the various philosophical problems with historical-philology, the ongoing usefulness of HALOT and BDB indicates that the approach does bequeath us much that is valuable.

Structuralists faulted the historical-philologists on three fronts: diachronic analysis treats meaning anachronistically; language should be understood as a coherent system, not as an atomistic repository of cultural ideas; and the “psychological conception of meaning” assumed by historical-philologists invests lexemes with too much semantic weight.[26] Instead, the structuralist school handled the word-concept problem by imagining language as “a conceptual layer between the mind and the world.”[27] That is to say, language itself was the imposition, by the mind, of an ordered structure upon the disorder of reality. There is an intuitive truth here; however, it fails to address the question of what a linguistic “meaning” of, say, a car actually is and how it relates to a car in the real world.[28] Barr himself acknowledged that structuralism had a “lack of attention to ‘meaning.’”[29] Ultimately this is because, for all its useful correctives to historical-philology, and right attention given to language as a coherent system, structuralism itself cannot solve the word-concept problem.

An influential development from structuralism was Chomsky’s generativist semantics. Chomsky took structuralism to its logical conclusion by asserting that language could be understood and treated as pure syntactical system apart from the world of referents and meanings.[30] This ushered in a raft of pragmatic (usage-based) approaches to semantics, among which is the corpus-based approach.

Corpus linguistics “considers the analysis of actual linguistic behaviour to be the ultimate methodological foundation of linguistics.”[31] Linguistic data sets are treated exhaustively, with specific criteria in view, such as collocation, governing verb, and so forth. The move by many biblical scholars to defend their conclusions by statements such as “having considered every verse where . . .” is a corpus-based approach. The strengths of this are obvious. It is a more scientific approach to explicitly delineate data sets for study and then develop theories that account for every occurrence of a phenomenon within such a data set. This gives the work of biblical scholars a far less subjective character.

Yet to claim that such analysis of usage is all there is to say about meaning goes too far, as Clines does in his Dictionary of Classical Hebrew, which confidently claims to be unique in having “a theoretical base in modern linguistics” due to its methodological commitment that “the meaning of a word is its use in the language.”[32] Muraoka severely criticizes this claim.[33] Ironically, by defining lexemes purely syntactically, the post-structuralist trajectory has come full circle in DCH by again equating word and concept.

If problems are generated by equating word and concept (historical-philology, TDOT), using word to define concept (structuralism, Barr), and separating them (generativism, DCH) then how can they be held together?

Cognitive linguistics approaches the question from a new direction. Rather than analyzing language and drawing conclusions about concepts, it instead begins by examining how humans conceive concepts and, as a result, how language functions. Word and concept can be equated, so long as “concept” is properly understood first. Cognitive linguistics as yet lacks a fully integrative theory, but despite its messiness its strengths are an interpretive power in explaining real-world linguistic phenomena, and the ability to integrate all previous avenues of linguistic investigation under one head.[34]

We therefore end our introduction to Hebrew linguistics and linguistic theory by making mention of the Semantics of Ancient Hebrew Database project. The SAHD describes lexemes by addressing all the possible avenues of data for lexical semantics: root and comparative material, formal characteristics, syntagmatics, translation in the versions, lexical/semantic fields, and exegetical observations.[35] One can see various tweaks that could be made to the model, such as a more sophisticated description of lexical fields using radial networks; nevertheless, it is extremely accomplished. While the SAHD does not explicitly subscribe to any one linguistic theory, it is encouraging to note that the need to draw from all the linguistic methodologies of the past is not unique to the present study but has also been recognized practically by Hebrew lexicographers.

The most immediate implication of cognitive semantics on the tripartite division will be the way it undercuts common critiques of the doctrine by giving a more sophisticated account of the way human beings perceive categories. By using it as a tool to analyze Reformed descriptions of the doctrine, I hope to demonstrate that these descriptions are far more sophisticated than most recognize, and that the doctrine in its classical form is theoretically defensible using this methodology. Let us, therefore, introduce five concepts from cognitive linguistics with this purpose in mind.

First, we introduce a cognitive semantic approach to synonymy. This is a key topic in the field of lexical semantics, and given that the tripartite division is effectively a discussion of Hebrew synonyms, it is effectively the central topic of this article. We then introduce the concept of lexical field. Lexemes cannot be understood in isolation alone (semasiology), but must be understood in relation to other words (onomasiology). Next, since language “is not a separate and independent module of the mind,” and linguistic meaning “is not separate from other forms of knowledge of the world that we have,” cognitive frame is introduced as the semantic structure underlying lexical fields.[36] Finally we describe schematicity and prototypicality, two concepts which are used to understand and describe semantic categories and the way they relate to each other.

2. Synonymy

Synonymy is a frequently discussed subject in lexical semantics and serves as a useful introduction to the complexities involved. Geeraerts provides a simple description: “‘Synonymy’ is a relationship of semantic identity, either between readings of a word or between words.”[37] While this appears straightforward, some basic examples reveal it is not.

Cruse points to the lexemes Alsatian and spaniel. Despite their significant semantic overlap—both are breeds of canine with European ancestry—the lexemes are not synonymous. Instead they are implicitly contrastive, both being hyponyms (sub-classes) of dog.[38] Synonymy also exists in degrees; to borrow another example from Cruse, the lexemes settee and sofa are intuitively more synonymous than brainy and shrewd.[39]

Furthermore, synonymy of lexemes under certain circumstances can mask more complicated semantic relationships. The taxonomical relationship between the hypernym (super-class) dog and its hyponym spaniel is to be distinguished from the part-whole relationship between the holonym (whole) can and the meronym (part) beer.[40] Yet despite the very different relationships between these two pairs, in many speech-acts the lexemes can be substituted for the other in a similar way, that is, without any difference in the denoted referent and with minimal difference in connotation. Under certain conditions, then, both a hyponym and a holonym can behave like synonyms, and in doing so their taxonomical or part-whole relationships may not be obvious from their syntagmatic or pragmatic use alone. For example, dog may be substituted for spaniel in the sentence, “I bought the kids a spaniel,” and similarly can may be substituted for beer in the question, “Do you fancy another beer?” Is it clear from these instances that dog/spaniel is taxonomic, while can/beer is meronymic? We can mentally picture different looking animals and intuitively group them together as dogs, and we can imagine ourselves performing the familiar action of pouring beer out of a can. It is knowledge like that which naturally leads us to substitute dog for spaniel and can for beer under certain conditions.

All of the above examples point to the need to understand the real world concepts that lie behind the lexemes, and why structuralists had to spend “a great deal of time devising schemes for classifying different kinds of meaning relation.”[41] They wanted language to be understandable as an autonomous system if it was to function as a kind of “simplified” model of reality, yet this meant that all the complexities and nuances that arise from sense relations between things in the real world had to be classified. Over time this simply proved to be impossible. Synonymy, hyponymy, and meronymy are first cognitive realities, which are then expressed linguistically. It is for this reason that absolute synonyms, lexemes that semantically correspond at every point identically, do not exist. If lexemes reflect our encyclopedic knowledge of the real world, then every lexeme can be distinguished at some level.[42]

The lexemes חק, מצוה, משפט, and others used by Reformed theologians to justify the tripartite division are considered by some to be synonyms (see the discussion in section 3). Therefore, our discussion will necessarily involve some analysis of the extent to which this assertion is valid.

3. Onomasiology

The distinction between semasiology and onomasiology is crucial for rightly handling the evidence for the tripartite division. Kurt Baldinger describes this distinction concisely: “Semasiology . . . considers the isolated word and the way its meanings are manifested, while onomasiology looks at the designations of a particular concept, that is, at a multiplicity of expressions which form a whole.”[43] Geeraerts describes this as “a difference of perspective: semasiology starts from the expression and looks at its meanings, onomasiology starts from the meaning and looks at the different expressions.”[44]

Both DCH and SAHD approach Hebrew lexicography from an onomasiological perspective. In particular, as we have observed, the SAHD includes a section on “lexical field” for each lexeme discussed. At its simplest, a “lexical field” is a set of semantically related words, and owes its origin to structuralist semantics.[45] Similarly, the term “semantic field” has been used to describe the network of concepts that underlie such a lexical field.[46] There are, however, various ways of understanding how such lexical and semantic fields are structured, and how they relate.

There are two key benefits of considering semantics from an onomasiological perspective. The first is in understanding diachronic change. Crucially, any semasiological change in an individual lexeme is a mechanism of onomasiological change. As the English word gay has become increasingly specialized to mean homosexual, other terms such as happy, chuffed, pleased, and so forth have either extended their meaning, acquired a new meaning, or been invented to express the concept of emotional positivity.[47] The lexical field responds to the change in one lexical item, to result in semasiological change in many. Considering diachronic change purely from the perspective of semasiology, however, does not account for this onomasiological development.[48]

Secondly, a consideration of onomasiological processes and structures presents a method for connecting pragmatics, the actual use of language, with meaning. From an onomasiological perspective we can ask the question, “Why use this word here?” When lexemes are understood from the perspective of their relationship to the rest of the lexical field, such pragmatic choices become real decisions about expressing nuances in denotation or connotation.

4. Cognitive Frame

As we have seen, linguistic phenomena such as synonymy cannot be explained without reference to the real world concepts to which lexemes point. Take, for example, the noun hypotenuse. A student asked to draw a hypotenuse on a blackboard would almost certainly draw an entire triangle, even though a hypotenuse is only one line of the triangle. A single line, however, is not understood to be a hypotenuse unless it is part of a right-angled triangle. The concept hypotenuse is part of a whole “frame of knowledge” which includes angles, Pythagoras, sine and cosine.

Such “frames of knowledge” or “cognitive frames” are “structured sets of beliefs and expectations that direct cognitive processing, including the use of language.”[49] They are abstracted ideas about reality which are simple enough to orient the thinker so that the right set of concepts comes to mind to understand the topic under discussion, but flexible enough to account for the complexity of the real world.[50]

The lexical field is the set of linguistic tools—literally the vocabulary—that a speaker uses to describe the content of a cognitive frame. Take the lexeme knight which can convey at least two different senses to a modern English speaker based on the lexeme’s association with two completely different cognitive frames. When knight is collocated with a lexeme or concept from within either of these cognitive frames, such as rook or crusades respectively, the mind instantly identifies what sense of knight is meant.

5 And 6. Schematicity And Prototypicality

If cognitive semantics could be cut open and dissected, its beating heart would be a concern for how human beings think about categories. We end our brief introduction of relevant concepts from cognitive semantics with two descriptive models of how human beings understand categories: schematicity and prototypicality.

Classical “Aristotelian” categories are defined by a necessary and sufficient list of attributes. The attributes are necessary for, and therefore inclusive of, every member of the set, yet simultaneously sufficient to exclude every non-member of the set.[51] Such Aristotelian discreteness has two problems. First, it fails to recognize the variety of internal structures within categories. Secondly, it fails to recognize the reality that many categories have fuzzy, non-discrete boundaries.[52]

Prototype theory recognizes that most categories exhibit four features, to a greater or lesser extent:[53]

1. Some members are more typical of the category than others.
2. The category’s edges are blurred.
3. The category cannot be defined by necessary and sufficient conditions.
4. Category members exhibit a family resemblance, and may form clusters.

Examples of prototypicality are not hard to find. To borrow several examples from Geeraerts: A robin is a “typical” bird. Penguins are not typical because they cannot fly, but there is no question that they are still birds. Yet the ability to fly still feels like a core attribute of what a “bird” is. The category fruit is even more prototypical in that it is impossible to define discretely. Typical attributes include sweet, fleshy, and containing seed. Many, however, struggle to classify olives or coconuts, and often mistake tomatoes for being vegetables, which feel far less fruity than strawberries even though strawberries do not contain their seed![54]

Categories relative to human relationships exhibit even less discreteness. Take, for example, the terms Evangelical, Reformed, or Anglican. While these are impossible to define discretely in an Aristotelian fashion, they continue in common parlance. Prototypical structures may also exist alongside one another in a taxonomy. Tax law, corporate law, and criminal law are all widely used hyponyms of U.K. law in legal circles, but these categories are not discrete, containing much overlap.

This last point leads us to schematicity. Whereas we might think of prototypicality as a flat network of members, with more typical members near the middle and less typical members further out, schematicity is the recognition of taxonomical hierarchy. One can think of birds at the specific and individual level, such as “a robin” or “the penguin we sponsor in the zoo.” The act of generalizing, or abstracting to a less specific level, such as “insectivorous birds” or “birds in captivity,” is schematization. The specific instances of an abstract schema “are called elaborations or instantiations or subcases of the schema.”[55]

Having introduced the key concepts of synonymy, onomasiology, cognitive frame, schematicity, and prototypicality, we now move to examine Reformed descriptions of the tripartite division of the Mosaic Law using these concepts.

II. Reformed Descriptions Of The Tripartite Division Examined From A Cognitive Linguistic Perspective

I am contending that the tripartite division has been rejected by many evangelicals too quickly, without being properly understood in its best articulation by the Reformed divines of the first two centuries of the Reformation.[56] Here we aim to employ the above introduced concepts from cognitive linguistics as a lens with which to sharpen our understanding of these classical descriptions of the doctrine.

First, we examine the lexical justification for the tripartite division, and hence the implications of synonymy. Second, we shall observe that some writers began to develop a recognition of the onomasiological dimension of the lexical field. Third, we shall observe how writers’ descriptions of the categories exhibit both schematic and prototypical features. Finally, we shall observe the presence of a universal cognitive frame which underpinned Reformed descriptions of the tripartite division.

1. Synonymy And Lexical Distinction

Turretin, like most Reformed theologians, sees lexical evidence for the tripartite division from “the diversity of the names by which [the Law] is designated in the Scriptures.”[57] He sees three Hebrew lexemes used to refer to the three categories: מצוה for moral, חק for ceremonial, and משפט for judicial.[58] In the Septuagint, this distinction is retained by using ἐντολή for moral; δικαίωμα for ceremonial; and κρίμα for judicial, though Calvin critiques the use of δικαίωμα by the Old Greek.[59] They see the distinction also expressed in the lexical field of the NT, for example, Col 2:14 speaks of the ceremonial law because Paul uses the term δόγμα.[60] Note that it is this lexical data which regulates Calvin’s application of the tripartite division when reading Paul. Calvin repeatedly insists that Paul’s argument throughout Gal 2:15–4:31 is in reference to the entire Law,[61] but acknowledges that the occasion (Gal 2:1–14) and application (Gal 5:1–6:18) of Paul’s argument are specifically about Jewish ceremonies. Why? Because in Gal 5:6 “the word Circumcision is put for ceremonies.”[62] Calvin will only acknowledge the presence of the tripartite distinction in Paul when there is an objective reason to do so. Whether Calvin is right to do this or not, the point is that his application of the tripartite distinction is not subjective but grounded in his understanding of what words mean. This is not the application of “an a priori grid”[63] upon Paul, as Carson claims, but exactly what Carson applauds Seifrid for doing when he imports “creational categories” into righteousness in Rom 1:17 because these are the associations righteousness has in the OT.[64]

Furthermore, the Reformed divines do not always assume that legal terms refer to the tripartite division, but display a sophisticated understanding of referentiality. Turretin, for example, acknowledges that the lexemes do not convey their specific nuance in every context but that sometimes “these words are synonymous and used promiscuously (Ezek 5:6; 20:11, 16, 18).”[65]

2. Onomasiology/Lexical Field

The most sophisticated lexical treatment of the terms is to be found in Witsius, who himself credits his observations to Beza. Witsius goes further than Turretin by arguing that while the terms are ordinarily synonyms when used alone, when used together they convey their lexical nuance.[66] This observation is, effectively, an onomasiological one, as the semantics of individual terms are being considered in light of the whole lexical field. Furthermore, it implies that the division is not necessarily even part of the internal semasiological structure of the lexemes themselves, which may explain why several older studies on the meaning of law synonyms (undertaken from an historical-philological perspective) failed to distinguish them semantically.[67]

A similar phenomenon to Witsius’s claim occurs with the above-mentioned lexemes settee and sofa. When the lexemes are used independently they are virtually absolute synonyms; only in very specific contexts would their nuance in meaning be distinguished. However, when used together settee may be intuitively felt to have an archaic or formal feel compared to the more informal sofa. They thus imply a difference in connotation; yet, given that the semantic knowledge behind the lexical field is encyclopedic, note that this connotative difference entails a degree of denotative difference![68] At the heart of cognitive schema are images, mental representations which are typical of the schema.[69] The mental image conjured by settee will be different from that conjured by the more comfortable sofa.

While writers like Beza, Turretin, and Witsius acknowledge that law lexemes do not universally and consistently refer to the divisions of the Law, their recognition of the semantic complexity actually adds, rather than removes, authenticity to their case. Their recognition reflects the onomasiological nature of language, which four hundred years later we can now describe with technical detail.

3. Prototypicality

While the Reformers are sometimes criticized that their retaining of the tripartite division is a hangover from the Aristotelian categories of medieval Catholic theology, ironically the most common critique of the tripartite division is precisely along Aristotelian lines. This is Sanlon’s second point, that the “question of what exactly goes into each division is impossible to settle,”[70] or as Wright puts it, individual laws “don’t fall neatly into those categories.”[71]

Furthermore, the acknowledgement of this by Reformed theologians is often accused of being internal inconsistency. Take, for example, Calvin’s observation that “ceremonial and judicial laws pertain also to morals. For the ancient writers who taught this division, although they were not ignorant that these two latter parts had some bearing upon morals, still, because these could be changed or abrogated while morals remained untouched, did not call them moral laws.”[72] Similarly, WCF 19.3 concedes that the ceremonial laws “partly, [hold] forth divers instructions of moral duties.”

The problems with this Aristotelian critique will be obvious from our previous discussion of prototypicality which recognizes that cognitive categories are not necessarily discrete sets. It is again an authenticating sign of the view of the divines that their descriptions of the biblical data appear centuries later to fit with the very common cognitive phenomenon of prototypicality. Fuzzy boundaries at the edge of categories do demonstrate that the tripartite division cannot be defined with Aristotelian discreteness, but like the categories of modern law such as tax, corporation, and criminal, such exceptions do not rule out the existence of the category altogether.

4. Schematic Network

The way the divines describe the relationship between the different types of law reveals a schematic network. For example, take Wollebius: “The ceremonial and political law is ancillary to the moral law.”[73] Bullinger describes the ceremonial and judicial laws as being “added” to the ten commandments “to expound and confirm them therewithal.”[74] Ursinus is more specific. He affirms that ceremonial and judicial laws “do, indeed, agree with the Decalogue, but cannot be deduced from it by necessary consequence.” This is because they “are subservient to it, as certain specifications of circumstances.”[75]

In other words, while we might distinguish the three categories moral, civil, and ceremonial, they do not exist at the same taxonomic level. The civil and ceremonial laws are (in cognitive terms) instantiations of the moral law for particular purposes in particular circumstances. This is why the moral law is eternal, but the ceremonial and judicial laws are different both before and after the existence of the Israelite theocracy.[76]

Yet many divines go further. While it is common shorthand to equate the moral law with the Decalogue,[77] when the Decalogue is examined in more detail many recognize that it is itself an instantiation of God’s character. For Ursinus the moral “agrees in every respect with the Decalogue, and may also, by necessary consequence, be deduced from it,” and he goes on to note that there is a difference “between the moral law, the natural law, and the Decalogue.”[78] We can distinguish the moral law and the Decalogue using Witsius’s statement about the latter being “published . . . to the Israelites” which means that it has “annexed some reasons, which, to the letter, were peculiar to them alone.” According to Witsius, these reasons are an “analogy” and have “mystical signification.”[79]

At this point the Decalogue itself sounds more and more like an instantiation of a higher taxonomical schema. For Bullinger the first commandment contains mystery;[80] the fourth commandment “is ceremonial, so far forth as it is joined to sacrifices and other Jewish ceremonies, and so far forth as it is tied to a certain time.”[81] The promise of the fifth commandment of long life in the land “properly belong[s] to the Jews.”[82]

This becomes explicit in Turretin, for whom the “eternal and archetypal law in God” is impressed onto man’s conscience as the natural law and inscripturated as the moral law.[83] The Decalogue is the ultimate historical expression of the moral law, but it is still an historical expression, and thus the foundation of Israel’s ceremonial and civil laws.[84]

The universal picture that emerges, often made explicit, is that the Decalogue (and indeed the idea of law itself) is an instantiation of love, the character of God.[85] Naturally, therefore, the greatest instantiation of this is given by God personally when he speaks to Israel directly and writes the tablets with his own finger. The civil and ceremonial laws are further instantiations of this Decalogue, revealed mediately through Moses.[86]

5. Frame Semantics

Given the above observations, if the evidence suggests that the doctrine of the tripartite division is a schema of cognitive categories, exhibiting features that the Reformers could describe but not label without the tools of modern cognitive linguistics, then crucial to assessing the validity of such a category schema is the cognitive frame that lies behind the lexical field. In other words, on what real-world basis is the tripartite division predicated? In the mind of ancient Israelites, how would the tripartite division be conceived, even if they would be unable to strictly classify every individual law?

For the Reformed divines, the tripartite division is a linguistic division that arises from a cultural/cognitive division between:

1. universal ethical norms which emanate from God’s character,
2. right order of worship within the people of God,
3. judicial law within the civil state.

This threefold distinction universally underpins all Reformed descriptions of the tripartite division of the Law. This is put most clearly by Turretin: “But the distinction appears principally from the nature of the thing and the office of the law. . . . The moral law regards the Israelite people as men; the ceremonial as the church of the Old Testament expecting the promised Messiah; the civil regards them as a peculiar people who in the land of Canaan ought to have a republic suiting their genius and disposition.”[87]

The cognitive frame, therefore, for the tripartite division is the socio-political world of the ANE. For the Reformed divines, the lexical distinction between moral, civil, and ceremonial is predicated upon this cultural distinction between ethics, worship, and the state. This distinction they consider to be intuitive and reasonable.

6. Implications

Given the amount of ground covered in sections 1 and 2, it is worth briefly taking stock at this point and drawing some preliminary conclusions and points of application. It was observed in the introduction that the doctrine of the tripartite division is particularly unpopular among evangelicals today. Indeed, as part of a nexus of neo-Lutheran ideas, it often goes hand in hand with an anti-law view of the Christian life.[88]

In the introduction, a clear commitment was made to tackle one argument against the tripartite division definitively: that “what exactly goes into each division is impossible to settle”[89] is not a valid argument against the doctrine. Even if the doctrine does not ultimately turn out to be supportable from the Scriptures, this argument—perhaps the most popular of arguments against the tripartite division—cannot be said to be valid if cognitive linguists are right about the reality of prototypicality structures.

Having examined the arguments of the Reformed divines carefully, I contend that they were simply too nuanced and too honest in their reading of Scripture for such an Aristotelian framework to be imposed upon them. This is not a surprise, as there seems such little indication that they have been allowed their own voice on the issue before the doctrine has been dismissed.

This observation, however, leads us to a relatively sober point of application for the church today: perhaps evangelicals should reflect on how we have often treated the views of our Reformed forebears. While I disagree with Westerholm and Carson and have argued that they have not paid sufficient attention to the historical defenses of the doctrine, it is true to say that they are both respectful and nuanced in their rejection of the tripartite division. Yet sadly such an attitude is not always emulated. Too often the tripartite division is ridiculed and the Reformed divines looked down upon for even contemplating such an obviously unscriptural position.

Hopefully these observations demonstrate something of the sophistication of the Reformed view. If the Reformed divines were mistaken, it was not for the somewhat simplistic reasons for which they are often criticized. This point is especially salient for those in Reformed denominations for whom the tripartite division is a confessional commitment. For example, serious reflection must be given to the pastoral implications of evangelicals who claim to “own” the Church of England because we stand in its theological tradition, but then deny, and even deride, the doctrines of that tradition, such as the tripartite division, without even properly understanding the case in its favor. For many non-evangelicals, such apparent inconsistency creates an unnecessary barrier toward their coming closer to the evangelical understanding of the faith.

Thus far we have demonstrated that some popular arguments against the tripartite division do not logically stand up when examined more closely from the perspective of cognitive linguistics and when the defense of the tripartite division has been placed on a firm methodological footing. Let us now demonstrate how this methodology can be applied to build a scriptural case.

III. A Snapshot Of The Biblical Evidence For The Tripartite Division From The Perspective Of Cognitive Semantics

In this final section, we seek to address the first of Sanlon’s points of contention with the tripartite division, that the Law is viewed by Scripture as “a single undividable entity.”[90] We have seen that a crucial plank of the biblical case for the division was the lexical distinction between חק, מצוה, and משפט. We observed that the Reformers’ descriptions of the doctrine were often more linguistically sophisticated than they knew. In particular, the nuanced positions of Witsius and Bucer recognized that these lexemes only refer to the tripartite division when collocated in certain circumstances.

Yet given the complexities of cognitive semantics that have been presented, some may be wondering how it is that the meaning of words can be established or described at all! Or perhaps, more specifically, if the popular “common sense” arguments against the tripartite division are invalidated by the assertion that humans conceive of categories in a more complicated and fuzzy way, then how would it be possible either to prove or falsify the existence of any distinction between categories of law in the OT?

In light of what we have discussed so far, there are three questions which must be addressed in order to evaluate the validity of the Reformed doctrine of the tripartite division:

1. Would the Israelites have had the appropriate cognitive frame(s) to distinguish between moral, civil, and ceremonial law?
2. From an onomasiological perspective, is there a semantic distinction between law “synonyms” (so called) in Biblical Hebrew?
3. From a semasiological and onomasiological perspective, can the distinction between these “synonyms” sometimes refer to moral, civil, and ceremonial law?

In the introduction, we considered several caveats about what could and could not be achieved within this section given the limitations of the present study. Let us now specify these more precisely.

First, we shall examine the evidence for a cognitive frame which supports the distinction, specifically from the Pentateuch. If it was demonstrated that bronze and iron age Israelites simply lacked the conceptual framework to distinguish ethics, civil institutions, and cultic practices, or that evidence for such a conceptual distinction did not exist, then naturally this would falsify the doctrine. Yet if evidence of a conceptual distinction does exist, then this demonstrates that there is a tripartite division of the Mosaic Law. What would remain unaddressed is the extent to which the Reformed divines were right to find that division throughout Scripture, and ultimately in the NT, in Paul, and even in Galatians(!) on lexical grounds. Given that the establishment of a cognitive frame underpins everything else, we shall dedicate the bulk of this section to this question. It is believed that the evidence presented here constitutes a good case for concluding that ancient Israelites had the necessary conceptual building blocks for the tripartite division to be plausible.

The second and third questions would require a monograph-length work to address. Our second task, therefore, shall be to present some preliminary evidence that the three lexemes חק, מצוה, and משפט can at times be semantically distinguished when they refer to law. If it were demonstrated that these terms cannot be semantically distinguished, or that the nature of their distinction cannot ever refer to the tripartite division, then this would not necessarily have a bearing on the existence (or not) of a tripartite division. It would, however, falsify the doctrine as articulated by mainstream Reformed divines, who universally saw lexical evidence for the doctrine and believed this lexical evidence had a bearing upon NT exegesis. However, if such a distinction can be demonstrated then the classically Reformed doctrine of the tripartite division remains plausible. Let us turn, then, to the main plank of this evidence, the presence of a supporting cognitive frame.

1. Cognitive Frame: Indications Of Conceptual Distinction Between The Categories Moral, Civil, And Ceremonial

We now introduce evidence that such cognitive frames could be evoked in the mind of Israelites to distinguish in regard to law between ethics, the judiciary, and the cult. We shall consider external evidence to Scripture from the cultures surrounding and in contact with Israel, and internal evidence from Scripture, specifically the Pentateuch.

External Evidence from Mesopotamian Culture. Of the comparative legal material from the ANE, Walton summarizes that there “are six major treatises currently extant (Ur-Nammu, Lipit-Ishtar, Eshnunna, Hammurabi, Middle Assyrian, Hittite).”[91] Sprinkle presents a simple summary of the implications of scholarship on ANE cuneiform law codes for understanding biblical law. He concludes that biblical law occasionally borrows from, and has overlapping categories with, cuneiform law codes and that this suggests “some sort of cultural influence,” yet the contrasts between them “indicate that biblical law and cuneiform laws have rather different ideologies.”[92]

Walton gives one such difference: “The mixing of cultic law into the legal genre is a unique phenomenon in Israelite formulation.”[93] Furthermore, he cites Finkelstein who states that “formulations of the proper procedures governing commerce and economic life” are absent from the biblical law codes, whereas cuneiform law collections “dwell on these themes at great length.”[94] This does suggest that ANE Israelites lived in a cultural context where a distinction was drawn between civil and cultic law, yet the evidence could also be understood as the biblical law codes’ rejection of this widespread distinction.

More significant is the function law treatises played, which Walton discusses by comparing them to the other types of treatise in the ANE, “All three (medical, legal, divinatory) serve as manuals that are compiled to teach principles to practitioners through paradigms.”[95] They hence “offer model justice.”[96] This is far from the modern model of civil law, which anticipates every possible scenario in statute. But even the common law model, where legal precedent is itself treated as a binding statute upon courts, is an inadequate analogy.[97] Joosten observes that of the extant documents from private law suits “there is not a single reference to a written code.”[98] The ANE king was not a modern day legislator, courts functioned without appealing to law codes either as legislature or binding precedent, and citizens did not live under the “rule of law” itself. Instead Walton describes king Hammurabi’s role as that of “expounding his judicial wisdom by compiling sample verdicts,” judicial wisdom endowed to him by the god Shamash.[99]

Notice that this is a schematic network: justice is instantiated in a law treatise, which is used by judges and scribes as a paradigm from which justice can be schematized and learned. Considered in this way, the distinction between justice and its instantiation in law is fundamental to the ANE mindset. The anti-tripartite argument which claims that a bronze- or iron-age ANE people would have made no distinction between law and ethics is an imposition of the modern Western post-Enlightenment mindset. In the West today, justice is defined by law.[100] In the ANE justice was exemplified by law.

Internal Evidence from Scripture. The Pentateuch presents ample evidence that it shares this ANE mindset. The administration of justice in Israel was not alien to the surrounding culture, but the reality of Israel’s cultural borrowing, universally recognized by OT biblical scholars, receives explicit confirmation in Exod 18 where Jethro advises Moses to appoint judges over the people. Here the החקים והתורות are used for precisely the same function as the ANE treatises: to train upstanding men so that they can render just judgment.[101]

The text of Exodus then distinguishes between two sets of laws. First is the Decalogue, described as הדברים, “the words,” in Exod 20:1, and spoken directly by God to the people.[102] It is this direct revelation of God to the people, speaking apodictic law in the second person, which is unique to Israel. The second set of laws are described as המשפטים, “the rules,” in Exod 21:1, and spoken to Moses who shall set them before the people.[103]

These two sets of laws are again distinguished in Exod 24. Moses tells the people כל־דברי יהוה וכל־המשפטים, “all the words of Yhwh and all the rules,” in Exod 24:3, and writes them down (Exod 24:4). These are called the “Book of the Covenant” in Exod 24:7. The tablets of stone contain התורה והמצוה, “the law and the commandment” (Exod 24:12), which Yhwh has written.[104]

What legal function do these two—clearly distinct—sets of laws play? Kline presents several convincing reasons for why the Decalogue should be understood as a suzerein-vassal covenant treaty between Yhwh and Israel.[105] This treaty establishes Israel as a nation. The Book of the Covenant is therefore best understood as the law treatise which Moses uses to train the judges he appointed in Exod 18.

After the building of the Tabernacle, the next stage in the narrative of the Pentateuch is Leviticus, which is itself a third legal corpus with instruction on cultic practice (chs. 1–17) and holy living (chs. 18–27). Similar to the ANE treatises on divinity, Leviticus feels like a manual for training priests. As with the Book of the Covenant, Moses acts as mediator, but much of the cultic material is specifically addressed to the priests (e.g., Lev 6:9, 20; 8:2; 9:1–2, etc.), and even when the people are addressed it is the priests’ actions that are in focus (e.g., Lev 1:5, 11, 15; 2:2; 3:2, 8, 13, etc.). The section on cleanliness further suggests that Leviticus functions as a manual for priests. Aaron is addressed alongside Moses and instructed how he is to teach the people cultic purity (e.g., Lev 11:1; 13:1; 14:33; 15:1). The Holiness Code does cover a range of topics, but the specific focus on sex in ch. 18, which is repeated in ch. 20, is again suggestive of a cultic context, especially as there is particular discussion about priestly holiness (ch. 21).

In light of the previous discussion about law in ANE culture, this evidence is very suggestive that these three treatises—the Decalogue, the Book of the Covenant, and Leviticus—played distinct roles in the life of Israel. They comprise a schematic network: Yhwh’s covenant (Decalogue) is unique in the ANE in being a direct instantiation of the justice, wisdom, and holiness of Israel’s deity. In turn, the Book of the Covenant and Leviticus are themselves instantiations of this covenant being mediated by Moses. The treatises appear to reflect three prototypical conceptual structures; hence, there is much overlap between all three of them, yet also an equally clear focus to each: the Decalogue is Israel’s covenant, the Book of the Covenant is a training manual for civil judges, and Leviticus is a training manual for priests.[106] It is not being claimed that these three treatises map onto the tripartite division in a simplistic manner; rather, they are indicative of a conceptual distinction between covenant, judiciary, and cult, which do appear to be three cognitive frames underlying the function of each of these law codes. Once the Aristotelian straitjacket imposed on the tripartite division by its critics has been shed, these three conceptual categories appear to emerge quite naturally from a simple reading of the Pentateuch in the light of ANE culture.

Furthermore, these three cognitive frames appear to be employed by the Deuteronomist in a different way to distinguish aspects of the Law for the purposes of applying it to periods of change in Israel’s history. Kline observes that Deuteronomy anticipates three states of Israel’s political existence: pre-conquest, pre-monarchy, and monarchical, and that the “laws of the Mosaic covenants were programmed from the outset for this succession of modifications in Israel’s polity.” We have seen the judicial system Moses appointed to function during his leadership in Exod 18. He appoints “a modified judicial system” to function upon entry into Canaan (Deut 16:18–17:13), and “for the more distant future, he incorporated . . . the law of the king” (Deut 17:14–20).[107] The Deuteronomist therefore treats aspects of the Law as subject to change, as they relate to the concept of Israel as a civil state. This underlying cognitive frame is assumed by these passages and is necessary for their application.

Kline’s observation can be extended to laws which must undergo adjustment in light of changes to the Tabernacle. Exodus 20:22–26 regulates the building of altars, yet Deut 12:5 anticipates the establishment of a permanent place where God shall be worshiped, and the Deuteronomist even recognizes that some food laws may therefore apply differently at the same time due to the distance some tribes will live from this chosen place of worship (Deut 12:20–27). Having thus affirmed that some laws will change over time, and that the Law will apply differently in different places at the same time, the Deuteronomist sees no contradiction between this and stating immediately afterwards that the entire Law must be kept (Deut 12:28).

As an aside, this throws light on those passages in the NT which insist that the Law must be obeyed in its entirety.[108] They do not necessarily mean that there are no distinctions between aspects of the Law, nor that obeying the Law in full might mean different things at different times and places.

In summary, while much of the Law remains unchanged, some aspects of law must change alongside developments in Israel’s civil structures, and other law must change alongside developments in Israel’s cultic institution. The Deuteronomist even recognizes that the Law needs to be applied differently in different places. Indeed, Deuteronomy even anticipates a time when the Israelites are physically separated from the Temple and the land, yet still considers it possible for them to “return to the Lord your God . . . and obey . . . all that I command you today.”[109] This statement borders on the ridiculous if the Deuteronomist did not have a conceptual framework of law where its demands were reasonably adjusted given the inability to worship at the Temple or conduct life in the land. The evidence suggests that such a cognitive framework is absolutely necessary to make sense of the Law’s own internal statements.

2. The Existence Of Distinction Between The “Synonyms”

Having established that there is a good case that ancient Israelites had a conceptual grid which would have enabled them to distinguish between aspects of the Law along moral, civil, and cultic lines, we now move on to consider the evidence for whether this tripartite division fits the classically Reformed description of the doctrine, that is, whether the lexemes חק, מצוה, and משפט can be used to indicate the distinction. Many claim that there is no semantic distinction between them at all, as various nineteenth-century studies from a historical-philological perspective have concluded.[110] Yet similar studies in the German tradition have concluded that semantic distinctions do exist, but these are lost when the lexemes are collocated, strikingly the exact opposite opinion to that of the Reformed divines!

This latter conclusion is to claim that the various words for law are hendiadyctic, that is, “one meaning is expressed by means of two words.”[111] When Ringgren follows Hentsche in asserting that lists of law lexemes in Deuteronomy together refer to the entire legal corpus with no legal distinction, they are effectively claiming that hendiadys can occur with three and even four synonyms![112] The purpose given for such redundant repetition is rhetorical effect. Yet there is no evidence that this phenomenon has precedent anywhere else in classical Hebrew other than in lists of law terms in works clearly dependent upon Deuteronomy. Instead, Braulik is more careful to assert only that חק and משפט lose their semantic distinction when collocated.[113]

These claims of absolute synonymy between the law lexemes, which at times begin to sound syntactically fanciful, appear even more questionable when the lexical data is analyzed from the perspective of cognitive linguistics. The following is a brief presentation of various avenues of investigation from such a perspective which begin to indicate that such a distinction must exist between them.

Size of Lexical Field. There are at least ten individual lexemes used to refer to law or laws in OT Hebrew: אמר, דבר, חק, מצוה, משמרת, משפט, ספר, עד, עדות, and תורה. Three of these have the concept “law” at the core of their semantic structure (חק, מצוה, and תורה).[114] The sheer number of lexemes is suggestive that they encode some useful semantic distinction.

Onomasiological Diachronic Analysis. We noted above that absolute synonyms do not exist from an onomasiological perspective because absolute synonymy amounts to lexical redundancy. Cruse observes that even if absolute synonymy were to exist, it would not do so for very long:

If the relationship were to occur, it would be unstable. There is no obvious motivation for the existence of absolute synonyms in a language, and one would expect either that one of the items would fall into obsolescence, or that a difference in semantic function would develop.[115]

The implications of this are highly significant. Whatever perspective is taken on authorship and redaction of the OT, these terms are collocated for several hundred years at least. The lexemes חק, מצוה, and משפט are collocated in Lev 26:15, attributable to P, the antiquity of which is argued by Milgrom,[116] and in Deut 6:1, which many consider Josianic or at least pre-exilic. חק, מצוה, and תורה collocate in Gen 26:5. מצוה and תורה collocate in Exod 24:12. While the vagaries of post-Documentary Hypothesis criticism are hard to pin down, this is at least reasonable evidence that these lexemes were collocated pre-exile. If a traditional view of Mosaic authorship is assumed, then these texts would be dated even earlier, perhaps as far back as the twelfth or thirteenth century. The lexemes then collocate in texts as late as 2 Kgs 17:37; 2 Chr 19:10; Neh 9:13; 10:30; and Mal 3:22, the latter being dated well into the post-exilic period.

Even if these lexemes were absolute synonyms in the earlier texts, one would at least expect them to have developed a functionally useful semantic distinction after several hundred years of use.

Syntagmatic Analysis of Lexical Collocation in Deuteronomy. The case for hendiadys when the lexemes are collocated in Deuteronomy is supported by the claim that the Deuteronomist is listing lexemes to rhetorically express completeness.[117] In actual fact the Deuteronomist is extremely partial to the use of lists for this purpose. In every other instance, however, the Deuteronomist creates a sense of wholeness of an entire category by listing meronyms within that category.[118] Deuteronomy 5:14, for example, emphasizes that the Sabbath universally applies by listing “you . . . your son . . . your male and female slave . . . your ox . . . your donkey . . . any of your livestock . . . or the resident alien in your towns.” Sometimes the overarching category is left unstated, and the list of meronyms functions to illustrate what is meant, such as the command in Deut 26:12 that tithe be given to “the Levites, the aliens, the orphans, and the widows,” which is not exclusive but clearly implies all who are in need.[119] Lists such as these abound.[120]

If each of these lists creates the rhetorical force of completeness by listing the contents of a category, then why should the lists of law lexemes necessarily be hendiadyctic when they refer to the Law in its totality? In fact, are not these parallels suggestive that such lists of law lexemes express completeness precisely by listing meronyms of Israel’s Law?

Occasionally the Deuteronomist uses lists not to describe wholeness, but to draw attention to a quality or attribute of a category, such as a city’s defenses,[121] or the materials used to construct idols.[122] Three of these lists, referring to Moses’ experience of leadership,[123] the descriptions of Sinai,[124] and Yhwh’s power in saving Israel,[125] may contain hendiadys between two items within the list; nevertheless, they are still composed of meronyms of a broader category.

The verse commonly used to justify the tripartite division by the Reformed divines is Deut 6:1. This verse contains a singular noun (מצוה) followed by plurals (החקים והמשפטים) which stand in the same grammatical relation to the rest of the clause. Comparing lists with this syntagmatic pattern in Deut 2:34; 3:6; 5:31; 6:1; 11:1; and 31:12 is even more suggestive.[126] Half of these occurrences are strings of law “synonyms” (Deut 5:32; 6:1; 11:1). The other three cases are clearly some form of loose apposition,[127] where the string of plurals is something akin to an apposition of material:[128] “every city [consisting of] men, women and children” (Deut 2:34; 3:6), “the people [consisting of] the men, the women and the children” (Deut 31:12).

Overall, seen from a syntagmatic perspective, this data is highly suggestive that lists of law words in Deuteronomy are lists of meronyms, and that Deut 5:32; 6:1; and 11:1 may possibly express a more detailed relationship between them, such as “the commandment [consisting of] the statutes and the judgments.”

This evidence supports the possibility that the semantic nuances of the lexemes do sometimes point to the tripartite division, even if this cannot be established definitively in a work of this size. Despite the caveats described, this constitutes real evidence for the tripartite division, in its classically Reformed form, that cannot be quickly discounted.

The acknowledgement made several times that the present work cannot definitively or exhaustively establish the semantic content of the lexemes in question is not intended as a tacit admission that the present study achieves nothing or could carry no persuasive power at all! This data constitutes a good case that a tripartite division of the Law is derivable from Scripture, and that the classically Reformed expression of the doctrine is a possible, even probable, fit for that data.

IV. Conclusion

In section 1, we introduced cognitive linguistics by explaining how it sits as a linguistic theory in the context of developments in Hebrew semantics and lexicography over the last century. We then explored five key cognitive linguistic concepts relevant to an examination of the tripartite division of the Mosaic Law. In section 2 these five key concepts were then used as a lens with which to read the classical Reformed descriptions of the doctrine afresh. Seen through this lens, the doctrine in its classical form is far more sophisticated and nuanced than is often understood. This observation both places the case for the tripartite division on a sound methodological footing and counteracts the popular argument that the doctrine cannot be true because it is impossible to neatly categorize every law as either moral, civil, or ceremonial. Human categories are often prototypical, not Aristotelian, and this was recognized by the Reformed divines even without the aid of modern linguistics.

In section 3, this methodology was applied to Scripture, where it was demonstrated that there is a good case for concluding that ANE Israelites had the necessary cognitive framework to distinguish between moral, civil, and ceremonial law, and that such a distinction appears necessary from various passages in the Pentateuch. We also presented reasonable evidence to support the conclusion that Hebrew words for law do likely encode a semantic distinction in the type of law to which they refer. While this last point cannot be definitively demonstrated in a study of this size, the equivalent discussion by Braulik amounts to less than two pages in length.[129] In other words, at least in comparison to other material in the field, the limits of this study do not justify its conclusions being written off quickly.

Overall, then, we feel that a good case has been made for evangelicals to at least reconsider their views on the tripartite division. The evidence of the Pentateuch would appear to demonstrate that at least some form of tripartite division must be acknowledged in order to make sense of the Law’s historical and eschatological vision. A definitive proof of the doctrine in its Reformed expression has not been given, but some major objections have been overturned, and scriptural evidence has been adduced in its support from a sound linguistic basis. At this stage the semantic content of the relevant lexemes has not been definitively established, but there have been some hints along the way that if further work confirms the trajectory of the current study it has the potential to impact a variety of fields—such as the implications for Pauline studies that certain lexemes may reference particular aspects of biblical law.

This leaves us with the question of what direction future research must take. There is still much to do. The most pressing need now is for a comprehensive semantic study of the lexemes חק, מצוה, משפט, and perhaps תורה. Ideally such work should be organized and presented in the style of an SAHD entry, with some additional considerations such as the implications of parallelism and the description of semantic structure using prototype theory. A significant amount of work was undertaken toward this goal for the present study, but due to limits of space could not be included. After such semantic work, collocations of the lexemes must be analyzed to discern the nature of their semantic overlap, and the referents of each occurrence of the lexemes must be identified by applying a careful exegetical eye. Alongside all this, the above work on cognitive frame must be expanded, and the definition of moral, civil, and ceremonial law firmed up. Once all this work is completed, the findings must then be integrated with Ross’s defense of the tripartite division from a biblical theological perspective. Finally, the implications of this work must be brought to bear upon NT exegesis—after all, that is where the rejection of the tripartite division begins and ends for many.

Notes

1. Justin, Dialogue with Trypho (ed. Michael Slusser; trans. Thomas B. Falls; Washington, D.C.: Catholic University of America Press, 2003), 37-46 (chs. 23-30). Note that Justin distinguishes between ceremonial and moral law in order to justify the Christian view on continuity between the testaments, pace D. A. Carson, “Mystery and Fulfillment: Toward a More Comprehensive Paradigm of Paul’s Understanding of the Old and the New,” in The Paradoxes of Paul (vol. 2 of Justification and Variegated Nomism; ed. D. A. Carson et al.; Grand Rapids: Baker Academic, 2004), 429. Ptolemy’s description can be found in his letter to Flora as recorded by Epiphanius, Panarion (trans. Frank Williams; New York: Brill, 1987), 200-202 (1.33.5.1-6.6); for a summary see Johannes Quasten, Patrology, Vol. 1: The Beginnings of Patristic Literature (Utrecht: Spectrum, 1975), 261.
2. Augustine, Faust. 6.2 (NPNF 4:167-68). Augustine only explicitly distinguishes the moral and ceremonial categories here. Cranmer affirmed the division in art. 7 of the 39 Articles. John Calvin, Institutes of the Christian Religion (ed. John T. McNeill; trans. Ford Lewis Battles; 2 vols.; Louisville, Ky.: Westminster John Knox, 2006), 1503-4 (4.20.15).
3. Art. 28 of the Augsburg Confession (only distinguishes moral from ceremonial); art. 4 of the Smalcald Articles (distinguishes church ceremonies from the commandments of God); chs. 5 and 20 of the Scottish Confession of Faith; art. 7 of the 39 Articles; art. 84 of the Irish Articles; art. 25 of the Belgic Confession (only distinguishes moral from ceremonial); ch. 19 of the Westminster Confession of Faith (WCF); ch. 19 of the Baptist Confession of Faith; ch. 19 of the Savoy Declaration. The Heidelberg Catechism also applies the distinction in speaking of ceremonies as a subset of the Law in Lord’s Day 6, answer 19. Questions 168-70, 181 of the Genevan Catechism also imply the distinction. In summary, the four primary confessional traditions of the Reformation—Lutheranism, Continental Presbyterianism, English Presbyterianism, and Anglicanism—all at least confessionally require the distinction between moral and ceremonial aspects of the OT law. The latter of these two traditions confessionally requires the distinction between all three aspects of the tripartite division. Notably, the historic confessionally Reformed roots of modern-day Baptists and Congregationalists also bear witness to the tripartite division, even if these seventeenth-century confessions have largely fallen out of use today.
4. In 1987 Moo described this idea as one that “gains steady ground as one moves through the literature of the last decade” (Douglas Moo “Paul and the Law in the Last Ten Years,” SJT 40 [1987]: 287-307, quote appears on 294). This essay is critical for a number of reasons, as Moo both surveys the state of scholarship in the wake of the NPP and begins to draw together some of the theological concerns. He also calls into question the label “Lutheran orthodoxy” used to describe the emerging opponents of the NPP (206), which later gave rise to the introduction of the terminology “modified Lutheranism” or the term used throughout this work “neo-Lutheranism.”
5. Carson, “Mystery and Fulfillment,” 429; Stephen Westerholm, Perspectives Old and New on Paul: The Lutheran Paul and His Critics (Grand Rapids: Eerdmans, 2004), 437-38.
6. Douglas Moo, “The Law of Christ as the Fulfillment of the Law of Moses: A Modified Lutheran View,” in Five Views on Law and Gospel (ed. Wayne G. Strickland; Grand Rapids: Zondervan, 1999), 319-76.
7. The first edition of the Five Views was published under the title The Law, the Gospel and the Modern Christian in 1993. See the discussion in Walter C. Kaiser, Jr., “God’s Promise Plan and His Gracious Law,” JETS 33 (1990): 289-302.
8. Given that the OT was inscripturated under the Mosaic Covenant, in practice this tends towards an undermining of the entire OT to a greater or lesser degree.
9. “I will argue that the Mosaic law is basically confined to the old era that has come to its fulfillment in Christ” (Moo, “The Law of Christ,” 322-23; see also his methodological principles at the top of 324). A methodological commitment to the building of doctrine first and primarily by the NT is expressed by Tom Wells, “The Christian Appeal of New Covenant Theology,” 7-14; and the negative implications of this on the Mosaic Law as a source of ethics in the Christian life are unpacked by Fred G. Zaspel, “The Continuing Relevance of Divine Law,” 148-60; both essays can be found in Tom Wells and Fred Zaspel, New Covenant Theology (Frederick, Md.: New Covenant Media, 2002).
10. See Robert P. Lightner, “Theological Perspectives on Theonomy, Part 2: Nondispensational Responses to Theonomy,” BSac 143 (1986): 138-40.
11. Peter Sanlon, “The Curse of the Law,” http://beginningwithmoses.org/bt-articles/227/the-curse-of-law, 13 (accessed April 23, 2012).
12. Sanlon is also known to me personally, and while I disagree with his rejection both of the Mosaic Law and the role of “law” in the Christian life at all (a position which I believe is accurately described by the label “antinomian”) he stands as one of the best spokespersons for that tradition in his high regard for, and practice of, personal godliness.
13. Sanlon, “Curse of the Law,” 25.
14. For example, does Jesus fulfill every aspect of the Passover lamb in the same way? Or does he fulfill some aspects of it literally in his manner of dying (John 19:36) and other aspects of it eschatologically in the New Creation (Luke 22:16)?
15. Even though Sanlon’s third point shall not be explicitly addressed here, evangelical Anglicans such as he and I must at least wrestle with the fact that Sanlon’s assertion is in direct contradiction to our inherited tradition. See, e.g., the third question of Cranmer’s catechism which states that the Christian life involves a resolution to “keep God’s holy will and commandments, and walk in the same all the days of my life” (The Book of Common Prayer [enlarged ed.; Cambridge: Cambridge University Press, 2004], 289).
16. Philip S. Ross, From the Finger of God: The Biblical and Theological Basis for the Threefold Division of the Law (Ross-shire: Mentor, 2010).
17. We shall discuss Barr under §1.1 below. Porter, perhaps a little too dismissively, describes Barr’s insights as still “widely ignored in much language-related biblical research” (Stanley E. Porter, “Greek Linguistics and Lexicography,” in Understanding the Times: New Testament Studies in the 21st Century; Essays in Honor of D. A. Carson [ed. Andreas J. Köstenberger and Robert W. Yarbrough; Wheaton: Crossway, 2011], 19).
18. For the insight that a coherent narrative of linguistic theory hangs together upon this theme I am greatly indebted to Dirk Geeraerts, Theories of Lexical Semantics (Oxford: Oxford University Press, 2010).
19. James Barr, The Semantics of Biblical Language (Oxford: Oxford University Press, 1961), 291.
20. Ibid., 21.
21. Barr himself was clearly a structuralist; see ibid., 294-95.
22. See Geeraerts, Theories, 2-5 and 10-14. For an example of such an approach see Thorleif Boman, Hebrew Thought Compared with Greek (trans. Jules L. Moreau; London: Norton, 1970).
23. HALOT 1:1xix.
24. Muraoka also speaks positively about the ongoing usefulness of these tools even after a hundred years; see Takamitsu Muraoka, “A New Dictionary of Classical Hebrew” (review of David J. A. Clines, ed., The Dictionary of Classical Hebrew), in Studies in Ancient Hebrew Semantics (ed. Takamitsu Muraoka; Abr-Nahrain Supplement Series 4; Louvain: Peeters, 1995), 87.
25. Barr’s criticisms can be applied to the TDOT as well. The preface and translator’s notes to the first edition make it perfectly clear that the dictionary’s philosophy is the same as that of Kittel (TDOT 1:v-vii). The admission in the editor’s preface to vol. 15 that “the original conception of the Dictionary required continuous modification” is also a striking indication that there were indeed flaws in its underlying philosophy (15:xi)! In fact, some articles are so theologically fanciful that TDOT’s own editors felt led to refute them (15:612).
26. Geeraerts, Theories, 48-51, quote appears on 49.
27. Ibid., 51. In this respect Chomsky’s idea that language can be treated autonomously was perhaps less revolutionary and more evolutionary; cf. Stephen Ullman, The Principles of Semantics (2d ed.; Oxford: Blackwell, 1957), 43-65, where Ullman presents an autonomous understanding of words as a result of the natural trajectory of developments in linguistics. Note that the first edition of Ullman’s work predates Chomsky’s Syntactic Structures by six years.
28. For further evidence that meaning cannot be understood apart from considering the real-world referents of lexemes, see the discussion at §1.2 and §1.4 below.
29. Barr, Semantics of Biblical Language, 295.
30. Noam Chomsky, Syntactic Structures (2d ed.; New York: De Gruyter Mouton, 2002), 17. Famously, when Chomsky first released this seminal work on linguistics, he included no section on semantics at all.
31. Geeraerts, Theories, 168.
32. David J. A. Clines, ed., The Dictionary of Classical Hebrew (6 vols.; Sheffield: Sheffield Academic, 1993-2007), 1:14; emphasis added. References shall be abbreviated to DCH.
33. Muraoka, “New Dictionary of Classical Hebrew,” 89-93.
34. For this reason, I see numerous correlations between cognitive linguistics and discourse analysis, the latter of which has certainly begun to make a mark in most areas of biblical studies.
35. Takamitsu Muraoka, Introduction to Semantics of Ancient Hebrew (ed. Takamitsu Muraoka; Abr-Nahrain Supplement Series 6; Louvain: Peeters, 1998), xi. Before any dictionary entries were published, studies were undertaken to advance the field’s understanding of how to integrate and weigh such data as semantic change in Late Biblical Hebrew, the implications of the versions, text criticism, etc. (Muraoka, ed., Studies in Ancient Hebrew Semantics).
36. Dirk Geeraerts, “A Rough Guide to Cognitive Linguistics,” in Cognitive Linguistics: Basic Readings (ed. Dirk Geeraerts; New York: De Gruyter Mouton, 2006), 4-5
37. Geeraerts, Theories, 84.
38. D. A. Cruse, Lexical Semantics (Cambridge: Cambridge University Press, 1986), 266. Goddard defines hyponomy as “where the meaning of one word is included in another” (Cliff Goddard, Semantic Analysis: A Practical Introduction [Oxford: Oxford University Press, 1998], Kindle edition, loc. 302.
39. Cruse, Lexical Semantics, 265.
40. Geeraerts, Theories, 88.
41. Goddard, Semantic Analysis, loc. 307.
42. Cognitive linguistics can be seen as seeking to make pragmatics a fundamental part of the understanding of language, rather than simply an offshoot from “core” language study. As such, one method of extending the above discussion about distinguishability of synonyms would be to examine lexemes from the pragmatic perspective of conversational implicature; see Stephen C. Levinson, Pragmatics (Cambridge: Cambridge University Press, 1983), 97-166.
43. Kurt Baldinger, Semantic Theory: Towards a Modern Semantics (Oxford: Basil Blackwell, 1980), 278.
44. Geeraerts, Theories, 23.
45. Cruse, Lexical Semantics, 112-35.
46. Geeraerts, Theories, 53-70.
47. Acquiring a new meaning or introducing a new term to a language is called lexicogenesis.
48. Geeraerts, Theories, 41.
49. Ibid., 224.
50. See Charles J. Filmore, “Frame Semantics,” in Basic Readings, 373-400; and Alan Cienki, “Frames, Idealized Cognitive Models, and Domains,” in The Oxford Handbook of Cognitive Linguistics (ed. Dirk Geeraerts and Hubert Cuyckens; Oxford: Oxford University Press, 2007), 170-87.
51. Aristotle, “Posterior Analytics,” in The Complete Works of Aristotle (2d ed.; ed. Jonathan Barnes; 2 vols.; Princeton, N.J.: Princeton University Press, 1995), 1:161 (2.13.7-15). This may also be referred to as the “Classical Theory of concepts” (Barbara Lewandowska-Tomaszczyk, “Polysemy, Prototype, and Radial Categories,” in Oxford Handbook of Cognitive Linguistics, 144).
52. Geeraerts, Theories, 65-70.
53. I.e., the phenomenon of prototypicality is itself prototypical!
54. Geeraerts, Theories, 184-92.
55. David Tuggy, “Schematicity,” in Oxford Handbook of Cognitive Linguistics, 83.
56. I use the term “Reformed” here to refer to those strands of the Protestant tradition which draw in particular upon Calvin as their legacy.
57. Francis Turretin, Institutes of Elenctic Theology (ed. James T. Dennison, Jr.; trans. George Musgrave Giger; 3 vols.; Phillipsburg, N.J.: Presbyterian & Reformed, 1992-1997), 2:146 (11.24.2).
58. The Reformed divines are not consistent among themselves in using the plural or singular of the relevant Greek and Hebrew lexemes, which seems probably to be influenced by the number of the noun in the passage which is uppermost in their minds as they write. For the sake of simplicity I render all Greek and Hebrew words in their lexical, singular form throughout this article. The Reformed divines also use a variety of different Latin and English words as labels for the three categories. This is less likely to confuse the reader as the categories are still clearly distinguishable, and so I have allowed the discussion in this section to reflect this variation of language.
59. Calvin, Institutes, 810 (3:7). Calvin is tackling the claim of some Roman Catholics that the term “justifications” is evidence that the Mosaic Law did justify, which he attributes to a poor rendering of חק.
60. Calvin, Institutes, 365-66 (2.7.17). Cf. Turretin, Institutes, 2:148-49 (11.24.9-10); and John Owen, The Works of John Owen (ed. William H. Gould; 23 vols.; Edinburgh: Banner of Truth, 1965-1991), 2:212, who ascribe this to the ceremonial law, but don’t give a lexical justification.
61. See in particular his discussion of the meaning of “works of the law” at Gal 2:15. His primary argument here is that the context demands that this phrase means the entire Law; he returns to remind the reader of this repeatedly throughout Gal 2:15-4:31 (John Calvin, Calvin’s Commentaries [trans. William Pringle; 22 vols.; Grand Rapids: Baker, 1984], 21:67-69). In fact, Peter Martyr mentions that a repeated point of contention between the Reformers and the Papists was over the application of the tripartite division to passages dealing with justification in Paul (Peter Martyr Vermigli, Predestination and Justification [ed. and trans. Frank A. James III; Kirksville, Mo.: Truman State University Press, 2003], 115-17). This strongly indicates that the doctrine was not simply adopted in an unreflected way. Furthermore, it is striking that the Reformers chose to maintain the doctrine despite the fact that it would have been to their advantage in the Reformation’s single biggest point of debate if they had simply denied the distinction existed in Paul at all
62. Calvin, Commentaries, 21:152
63. Carson, “Mystery and Fulfillment,” 429.
64. D. A. Carson, “Summaries and Conclusions,” in The Complexities of Second Temple Judaism (vol. 1 of Justification and Variegated Nomism; ed. D. A. Carson et al.; Grand Rapids: Baker Academic, 2001), 537.
65. Turretin, Institutes, 2:146 (11.24.2).
66. Herman Witsius, The Economy of the Covenants Between God and Man Comprehending a Complete Body of Divinity (trans. William Crookshank; 2 vols.; Phillipsburg, N.J.: Presbyterian & Reformed, 1990), 2:167 (4.4.8). Note also that the Scottish Confession of Faith cites collocations of law synonyms in Josh 1:3; 23:4 as its proof text for the tripartite division (ch. 5).
67. J. van der Ploeg, “Studies in Hebrew Law,” CBQ 12 (1950): 248.
68. Cf. the discussion concerning the distinction between the German lexemes Stuhl and Sessel in Geeraerts, Theories, 66-68.
69. Image schema is an entire subject in itself, yet space did not permit an introduction to this important concept in the discussion of section 1. See Todd Oakley, “Image Schemas,” in Oxford Handbook of Cognitive Linguistics, 214-35; and Raymond W. Gibbs, Jr., and Herbert L. Colston, “Image Schema: The Cognitive Psychological Reality of Image Schemas and Their Transformations,” in Geeraerts, Basic Readings, 239-68.
70. Sanlon, Curse of the Law, 25.
71. Christopher J. H. Wright, Living as the People of God: The Relevance of Old Testament Ethics (Leicester: InterVarsity, 1983), 14.
72. Calvin, Institutes, 1503 (4.20.14). Note that to claim that there is moral within the ceremonial is an implicit acknowledgement that the category “moral” is one that transcends the Decalogue. We shall explore this further in the section on schematicity.
73. Johannes Wollebius, “Compendium theologiae christianae” in Reformed Dogmatics: A Library of Protestant Thought (ed. and trans. John W. Beardslee III; New York: Oxford University Press, 1965), 79 (14.1).
74. Henry Bullinger, Decades (ed. Thomas Harding; trans. H. I.; 4 vols.; Cambridge: Cambridge University Press, 1899), 3:220. Note that while Bullinger believes that all the ceremonial and judicial laws can be “referred” to the Decalogue in some way, he does not think this necessarily means that there is a like-for-like correspondence between every law and an individual commandment: “Some laws may be applied to more precepts than one of the ten commandments.”
75. Zacharias Ursinus, Commentary on the Heidelberg Catechism (trans. G. W. Williard; 4th American ed.; Cincinnati: Elm Street, 1888), 492.
76. Richard Hooker, The Works of That Learned and Judicious Divine, Mr. Richard Hooker: With an Account of His Life and Death (ed. John Keble; Oxford: Clarendon, 1876), 387 (3.10.4); Calvin, Institutes, 1503-5 (4.20.15-16).
77. Witsius, Economy of the Covenants, 2:162 (4.4.2); WCF 19.2-3.
78. Ursinus, Commentary, 490. This distinction between the Decalogue and the moral law may also be why Vermigli places them both side by side (Vermigli, Predestination and Justification, 116).
79. Witsius, Economy of the Covenants, 2:176 (4.4.32).
80. Bullinger, Decades, 1:218-19.
81. Ibid., 1:259. Bullinger’s view is identical to the one expressed in the Anglican homily on the keeping of churches, that the essence of the Sabbath commandment is that there should be order in religion, but the specific order outlined is part of the Old Dispensation. Here I disagree with Ross’s view of the Sabbath, even though I stand with him on the tripartite division; cf. ibid., 1:202. This is also Turretin’s view (Institutes, 2:8).
82. Bullinger, Decades, 1:287; also Turretin, Institutes, 2:8.
83. Turretin, Institutes, 2:12 (11.2.4-5). Turretin outlines the differences between the natural law and moral law in 2:6-7 (11.1.12).
84. Ibid., 2:85 (11.13.22); cf. 2:13 (11.2.21).
85. Nowhere is this idea realized more masterfully than in the order of Bullinger’s Decades. Bullinger ends the first decade with a sermon on love, which he then unpacks in the following two decades on law, an architecture Bullinger makes explicit at the beginning of the second decade: “The sum of all laws is the love of God and our neighbour; of which and every part whereof because I have already spoken in my last sermon, the next is, that now also I make a particular discourse of laws, and every part and kind thereof” (Bullinger, Decades, 1:193).
86. Witsius, Economy of the Covenants, 2:163 (4.4.3). Note that Bullinger has an interesting thesis that the angels revealed the civil and ceremonial law to Moses, and this also distinguishes them from the Decalogue (Bullinger, Decades, 1:212).
87. Turretin, Institutes, 2:146 (11.24.2); emphasis added. See also “this distinction, which is evidently founded on the things themselves” (Owen, Works, 17:481); and “the ark of the Lord was the centre and compendium of all the ceremonies” (Witsius, Economy of the Covenants, 2:391 [4.14.27]).
88. Even though many of the individuals who put forward a firm “anti-law” position would shirk from the Latin equivalent of that term because they still believe in the transforming work of the Spirit. Yet to define “antinomian” as “anti-holy” is as historically fair as defining “nomian” as “legalist.” To deny the role of the Mosaic Law in the Christian life is antinomian, to deny the role of “law” altogether is more so. This is not to deny that those who hold to such antinomian views do indeed strive for, and value, godliness, yet there will be pastoral implications of reconfiguring the shape of the Christian life. Pace Sanlon, Curse of the Law, 22.
89. Sanlon, Curse of the Law, 25.
90. Ibid., 25.
91. John H. Walton, Ancient Near Eastern Thought and the Old Testament: Introducing the Conceptual World of the Hebrew Bible (Nottingham: Apollos, 2007), 288.
92. Joe M. Sprinkle, Biblical Law and Its Relevance (Lanham, Md.: University Press of America, 2006), 45, 47.
93. John H. Walton, Ancient Israelite Literature in Its Cultural Context: A Survey of Parallels Between Biblical and Ancient Near Eastern Texts (Grand Rapids: Zondervan, 1989), 76.
94. Ibid., 77.
95. Ibid., 289.
96. Ibid., 290.
97. I am indebted to my good friend Stephen Moore for this observation.
98. Jan Joosten, People and Land in the Holiness Code: An Exegetical Study of the Ideational Framework of the Law in Leviticus 17-26 (New York: Brill, 1996), 19.
99. Walton, Ancient Near Eastern Thought, 295.
100. All the more so as ideology becomes increasingly democratized, and the definition of right and wrong increasingly becomes a function of the consent of the majority.
101. Exod 18:20-22. “Statutes and laws.”
102. Exod 20:1, 18-21.
103. Exod 21-23.
104. See also Exod 31:18; Deut 9:10. The implication of this is that they unmistakably come from God; cf. Exod 8:19; Luke 11:20.
105. Meredith G. Kline, The Structure of Biblical Authority (2d ed.; Eugene, Ore.: Wipf & Stock, 1997), 113-30. The form of the Decalogue resembles that of such ANE treaties. The two tablets are therefore likely to be two identical copies, as the custom was to make two copies and place them in the temple of the gods of both the suzerain and the vassal. Similarly the two tablets of the Decalogue were placed in the ark, which resided in Israel’s Temple, Exod 40:20; Deut 10:1-5. This is the purpose of the ark, Exod 25:16, 21; it is called ארון העדת, ”the ark of the testimony,” Exod 25:22; 26:33-34; 30:6, 26; 31:7; 39:35; 40:3, 5, 21; Num 4:5; 7:89; Josh 4:16. It is therefore also ארון ברית, “the ark of the covenant,” Num 10:33; 14:44; Deut 10:8; 31:9, 25-26; Josh 3:3, 6, 8, 11, 14, 17; 4:7, 9, 18; 6:6, 8; 8:33; Judg 20:27; 1 Sam 4:3-5; 1 Kgs 3:15; 6:19; 8:1, 6; 1 Chr 15:25-26, 28-29; 16:6, 37; 17:1; 22:19; 28:2, 18; 2 Chr 5:2, 7; Jer 3:16, and Heb 9:4; Rev 11:19 (ἡ κιβωτὸς τῆς διαθήκης).
106. In fact, given that priests played a quasi-judicial role in early Israel (cf. Deut 17:8-13) it is not surprising that there is overlap of content in some areas of the Holiness Code with the Book of the Covenant if, as we have suggested, these legal treatises both function as training manuals.
107. Kline, Structure of Biblical Authority, 103.
108. E.g., Gal 5:3; Jas 2:10.
109. Deut 30:2, esv.
110. van der Ploeg, “Studies in Hebrew Law,” 248.
111. E.g., Gen 3:16; Amos 3:10; Job 10:21 (Ronald J. Williams, Williams’ Hebrew Syntax [3d ed.; rev. and exp. John C. Beckman; Toronto: University of Toronto, 2007], §72). I am grateful for John Beckman’s kind correspondence with me on this topic.
112. TDOT 5:142.
113. Georg Braulik, “Die Ausdrücke für ‘Gesetz’ im Buch Deuteronomium,” Bib 51 (1970): 61-62. For a summary of Braulik’s findings see Ross, From the Finger of God, 355-56.
114. I.e., in the vast majority of cases the referent of these terms is law.
115. Cruse, Lexical Semantics, 270.
116. Jacob Milgrom, Leviticus 1-16 (AB 3; Garden City, N.Y.: Doubleday, 1991), 3-13.
117. Deut 4:45; 6:17, 20; 8:11; 11:1; 26:17; 30:16.
118. I.e., parts of a whole.
119. See also Deut 27:19.
120. E.g., all religious offerings that shall be offered and eaten at the appointed place, Deut 12:6, 11, 17; all the possessions belonging to a soldier in Israel, Deut 3:19; and in a plundered city, Deut 20:14; all the inhabitants of a household, Deut 5:14, 21; all in Israel, Deut 12:12, 18; 16:11, 14; 29:9-10, 17; all friends and relations, Deut 13:6; all which one desires, Deut 14:26; all kinds of interest, Deut 23:20; all produce, Deut 28:4, 18; 30:9; all economic produce, Deut 28:51; all creation, Deut 10:14.
121. Deut 3:5.
122. Deut 29:16.
123. Deut 1:12.
124. Deut 4:11; 5:22.
125. Deut 6:22; 7:19; 8:17; 11:2.
126. Deut 5:31 and 11:1 are slightly different in that the conjunction is placed after the first singular noun. This is probably the reason for the similar insertion of the conjunction in Deut 6:1 by some witnesses.
127. Cf. Joüon §131m.
128. Cf. Williams, Hebrew Syntax, §68.
129. Braulik, “Die Ausdrücke für ‘Gesetz’ im Buch Deuteronomium,” 61-62.