A certain philosophy regarding issues of sufficient data base, semantic situation, unaffected vs. affected meaning, synchronic priority, and structural priority governs the approach of this study.1 These topics along with issues of lexical elasticity and some problematic data are addressed below.
This research attempts to identify 200 clear examples of SPN constructions consisting of the target cluster, or to find every occurrence in Koine Greek. The goal expresses the belief that only hard data can substantiate semantic assertions for any of the six possible structural types in question. The data pool of an initial study using Accordance Bible Software 5.1 included the New Testament (NT), the Septuagint, and the Apostolic Fathers (AF).2 A search of these sources did not yield the sought number of examples. Consequently, Thesaurus Linguae Graecae #E was used to expand the scope of the search to include extra-biblical works from the first century B.C.3 But this yielded a total of 21 examples. Therefore, the data pool was expanded to include all texts from the Koine Greek period.4
This criterion simply addresses the issue of the environment within which practices or “rules” are said to hold true. Solutions for many engineering problems, static or dynamic in nature, require that certain boundary conditions be known and fixed. The same holds true for grammatical studies. Formal statements about meaning of certain syntactical structures must be expressed as only true for examples that parallel the semantic situation of the entity under review. Conversely, functional statements about the diverse structures which convey similar meanings must be expressed as only true for examples that parallel the semantic situation of the entity under review. A few elements which define a semantic situation are context, genre, figures of speech, and morpho-syntactical features. The desire to adhere to this consideration, preserving the same or similar semantic situation, contributed to several decisions regarding the scope of this study, namely regarding three topics: (1) arthrous proper nouns, (2) a narrowly defined verb, and (3) the mood of the equative verb.
Review of secondary sources revealed that it is not uncommon for a proper name to appear with an article—for example when the proper name has been previously mentioned, when its identity has been made clear from context, or when it is well known.5 If traditional etymological theories are correct, the article derives from the demonstrative pronoun and initially functioned deictically (singled out or pointed out). Nominalization and definitizing uses may have grown out as natural by-product, or secondary uses. In reality the article has many other uses. Many grammars rightly describe it as primarily a function-word. By nature, function words create a semantic relationship between the word to which it is related and the rest of its environment. In this sense, it affects the semantic situation significantly. Since a proper noun is already definite, an article related to it most probably works only as a function-word, adding one more semantic relationship to the environment. Several grammars called this “marking” the proper name. Consequently, this study will omit marked proper nouns so as to search for a pecking order between a proper noun and an articular noun in a relatively well fixed semantic situation.
Similarly, focusing the study on the εἰμί verb alone also helps to maintain the same semantic situation. In fact, there are other verbs that might have been examined since they can also yield SPN constructions. Verbs like γίνομαι and ὑπάρχω can function in the same way as the equative verb, εἰμί, within certain semantic situations. Their lexical domains certainly allow it and their predicates usually agree with their subject in case. The fact is that all equative verbs function somewhere on the gradient of “purely copulative” to “truly predicative.”6 Dotson Nelson uses the labels of “form-word” and “idea-word” to explain the difference, and concludes that a true copulative verb simply connects a subject to predicate with minimal to zero importation of meaning in the process.7
The initial study revealed that, unlike εἰμί, γίνομαι and ὑπάρχω never show up as copulatives between a proper noun and an articular noun.8 In other words, they more often function as “idea words” in such cases. This raises the suspicion that they might never truly function as “pure” copulas in an SPN construction. In this sense, should they appear in SPN construction in an expanded data pool, preservation of the semantic situation may be suspect.
Most grammarians will note the absence of verbless constructions in this study. Indeed, verbal ellipses do form SPN constructions as Winer observes, “Of the three constituent parts of a proposition, the subject and predicate are indispensable; but the simple copula is implied in the mere juxtaposition of the subject and predicate.”9 Several reasons precipitated the decision to leave verbal ellipses out of the study. First, the searches yielded too many matches for these types of search constructs. Second, and by way of consequence, these types of matches required too much interpretation on the use of the nominatives.10 Both of these make identifying this type of SPN construction practically cumbersome. Third, the grammars and cursory review of the search results for verbless constructions suggest that there is a low probability of finding any examples of our problem case from this sample pool. This is because the predicate is frequently anarthrous in these SPN constructions.11
Finally, the mood of the equative verb also affects the semantic situation. The previously mentioned preliminary study also revealed that in every case of the examples found, the verb was in the indicative mood. Therefore, in order to maintain the same semantic situation, as it is affected by the mood of the verb, the study only considers sets/combinations of the nominative anarthrous proper noun and articular singular noun with the 3rd person singular equative verb in the indicative mood.
“By ‘unaffected’ is meant the meaning of the construction in a vacuum—apart from contextual, lexical, or other grammatical intrusions. By ‘affected’ is meant the meaning of the construction in its environment—i.e., ‘real life’ instances.”12 Herein resides the crux of the hypothesis. What is the difference between the anarthrous proper noun and the articular singular noun in the nominative case? In an SPN construction, does one exist ontologically or merely phenomenologically? The study attempts to answer this question.
The Greek verb contains both subject and predicate, the simplest sentence represented by the word, εἰμί. Several grammarians agree that any “subject” in a Greek sentence stands in apposition to the pronoun ending of the verb. Dana and Mantey write, “The original function of the nominative was to lend more specific identification to the subject [expressed by verb ending] of a finite verb.”13 They continue the thought by saying, “Consequently when we express a noun subject of the verb, it is in apposition with the subject implied in the verb itself.”14 For example, Παῦλος ἐστιν ὁ δοῦλος can be translated “He, namely Paul, is the bond-servant.” The proper noun assumes the verb’s embedded subject and the articular noun asserts something about it. The question arises, can this sentence be translated “He, namely the bond-servant, is Paul?” At the unaffected level, it can be translated as such only if it can be shown that the proper noun asserts more than the articular noun. To put it another way, the translation is valid if it can be shown that the articular noun assumes more than the proper noun.15
Unfortunately, the validation process for this kind of question is somewhat problematic. Ontological statements regarding the “assumption” or “assertion” level of a proper noun or of an articular noun must “be made on the basis of carefully scrutinized and representative phenomena.”16 This is nearly impossible to do with convertible propositions. The previous example can be embellished to create a stand alone convertible proposition; “He, namely the author of Romans, is Paul.” But notice that it stands alone only to the degree that the mind of the reader is familiar with the intrusions “the author” or “of Romans.” It shows that the grammatical subject cannot be determined apart from phenomenological factors.
By way of further example, note the citations given as exceptions to the proper name priority rule in Wallace’s grammar.17 The nouns in ᾿Ιωάννης ἐστὶν ὄνομα αὐτοῦand in οὐχ οὗτός ἐστιν ὁ τοῦ τέκτονος υἱός;οὐχ ἡ μήτηρ αὐτοῦ λέγεται Μαριάμ trump the proper names as subjects.18 This is primarily due to intrusions, namely the αὐτοῦ in each verse. Furthermore, the lexical force of the anarthrous ὄνομα in Luke 1:63 also constitutes an intrusion to the construction under examination. Again, this shows that in a convertible proposition the subject is best determined from context.19
In summary, this study focuses on affected constructions rather than on unaffected ones. This is because convertible propositions are by definition largely influenced by phenomenological factors. Perhaps, this is more reason to proceed with extreme care and respect for structural priority.
A diachronic study will not be used in this research because it is more interested with the “syntactical phenomena embedded in the NT.”20 The original research aimed at finding a minimum of 100 or more clear examples of the target cluster (nominative anarthrous proper noun, the 3rd person singular equative verbs, and the articular singular noun in the nominative case). However, preliminary studies of the original sample pool only produced a total of twenty-one target clusters, fifteen of which were convertible propositions. The initial results support the project’s falsifiable hypothesis but do not meet the sufficient data base goal. Consequently, a full synchronic study was executed using TLG #E and other resources to search for the same target cluster in the Koine Greek period.21
“The starting point of [this] investigation will be the given structure from which [it hopes] to make semantic conclusions.”22 The target cluster under examination for this study consists of an anarthrous proper noun in the nominative case, an articular singular noun in the nominative case, and the 3rd person singular equative εἰμί, verb in the indicative mood. This basic set yields six “given structures” from which the study will make semantic conclusions. In order to keep the number of structural types at six, this study will omit cases where the proper noun is arthrous. “Marking” the proper noun with the article adds another variable to the analysis.
For the purpose of this study, terms are taken to be nouns, as opposed to being adjectives or participles, when they are so defined by Bauer’s or Scott-Liddell’s respective lexicons.23 Articular adjectives and participles functioning as substantives are not included as primary evidence for empirical substantiation.24 Furthermore, terms in biblical literature are treated as proper nouns only if they can be found in Young’s Analytical Concordance to the Bible.25 For the non-biblical literature, when indexes of names were provided at the end of any translated book, they too were consulted along with the lexicons. Place/thing names and proper names are labeled proper nouns, and treated as equivalent syntactical units.26
By way of further clarification, θεός is not treated as a proper name. Studies by Philip Harner, Paul Dixon, and Wallace have shown that θεός can span the semantic fields of indefinite, qualitative, or definite noun.27 It seems that this kind of lexical elasticity is not shared by most proper names. In order to ensure the “tightest” target cluster sense possible, this study does not consider SPN constructions consisting of an equative verb, an articular noun, and θεός.
Similarly, even though Paul uses Χριστός often as a name for Jesus, it is not technically a proper name.28 In conjunction with ᾿Ιησοῦς, it definitely carries the force of a proper name but this is not lexically driven. In the Septuagint, the word also functions as an adjective.29 And while it is more often used as a noun in the NT, it is in the insipient stages of becoming a proper name. In this sense, it is also more lexically elastic than most proper names. In order to preserve the integrity of the structural types, it seems best not to consider it as a proper name for the NT literature and to allow it for post-NT literature where context clearly shows that the author employs it as such.30
There are several passages which were close but did not qualify, formally speaking. For example, John 18:40 contains an article before the proper name, ἦν δὲ ὁ Βαραββᾶς λῃστής. If it belongs to the proper name for the reasons of re-introduction then the following noun is anarthrous, which disqualifies this datum.31 If the article belongs to the noun, then this is a peculiar structural type, VANPN. Several verses like these were discovered. However, they were excluded in order to preserve structural priority.32 Technically speaking, Mark 10:47 formally qualifies,ὅτι ᾿Ιησοῦς ὁ Ναζαρηνός ἐστιν. But due to the idiomatic nature of ὁ Ναζαρηνός (“of Nazareth” rather than “the Nazarene”), it is best to pull the subject out of the verb and place the articular noun in apposition to the proper name, “that it was Jesus the Nazarene.”33
1 Borrowed from Daniel B. Wallace, Greek Grammar beyond the Basics: An Exegetical Syntax of the New Testament (Grand Rapids: Zondervan, 1996), 1-7.
2 Roy Brown, Accordance Bible Software Ver. 5.1 (Vancouver: OakTree Software, Inc., 2004). This program uses Alfred Rahlfs’ text of the Septuagint and Nestle-Aland’s 27th edition of the NT text (Alfred Rahlfs, Septuaginta, vol. 1 & 2, 9th ed. [Stuttgart: Württemberg Bible Society, 1935 (repr. 1971)]: 1-941; Eberhard Nestle, The Greek-English New Testament, ed. Erwin Nestle, rev. and ed. by Kurt Aland and Barbara Aland, 27th ed. [Stuttgart: Deutsche Bibelgesellschaft, 1998], 1-886). Rahlfs’ text consists primarily of the important manuscripts Codex Vaticanus and Codex Alexandrinus, BA. It also provides the parallel Codex Sinaiticus, S, in places where there is significant disparity between BA and S. For the discussion in this chapter, LXX1 represents BA and LXX2 represents S.
3 Thesaurus Linguae Graecae Ver. #E (Los Angles: University of California, 1999). The latest update of the CD-ROM, version #E, was released in February of 2000. It contains 76 million words of text. The online version was released on April 2001. It contains 91 million words of text. This program uses the second edition of the United Bible Society NT text and Alfred Rahlfs’ Septuagint text (Eberhard Nestle, The Greek New Testament, ed. Erwin Nestle, rev. and ed. by Kurt Aland, Matthew Black, Carlo M. Martini, Bruce M. Metzger and Allen Wikgren, The Greek New Testament, 2d ed. [Stuttgart: Württemberg Bible Society, 1968], 1-895; Alfred Rahlfs, Septuaginta, 9th ed. [Stuttgart: Württemberg Bible Society, 1935 (repr. 1971)], 1-941). Rahlfs’ text consists primarily of the important manuscripts Codex Vaticanus and Codex Alexandrinus, BA. It also provides the parallel Codex Sinaiticus, S, in places where there is significant disparity between BA and S.
4 The sample pool from the original study yielded 3,810 matches and only seven qualified as target clusters functioning in SPN constructions. Since the goal of this thesis is to identify 200 examples, the scope was expanded to cover 700 hundred years. These produced 75,918 potential matches.
5 Edwin A. Abbott, Johannine Grammar (London: Adam and Charles Black, 1906), 57-58; Alexander Buttmann, A Grammar of the New Testament Greek, trans. Joseph Henry Thayer (Andover: W. F. Draper, 1876), 86-87; H. E. Dana and Julius R. Mantey, A Manual Grammar of the Greek New Testament (New York: Macmillan, 1957), 142-44; Basil Lanneau Gildersleeve, Charles William Emil Miller, and Peter Stork, Syntax of Classical Greek: From Homer to Desmosthenes, Reprint ed. (Groningen: Bouma's Boekhuis B.V., 1980), 229-42; Stephen A. Janssen, “The Greek Article with Proper Names in Matthew” (Th.M. thesis, Dallas Theological Seminary, 2003); Stephen H. Levinsohn, Discourse Features of New Testament Greek: A Coursebook (Dallas: Summer Institute of Linguistics, 1992), 98-99, 217; Dotson M. Nelson, “The Articular and Anarthrous Predicate Nominative in the Greek New Testament” (Th.D. diss., Southern Baptist Theological Seminary, 1944), 5; Herbert Weir Smyth, Greek Grammar, rev. by Gordon M. Messing (Cambridge: Harvard University Press, 1956), 289-91; A. T. Robertson, A Grammar of the Greek New Testament in the Light of Historical Research, 4th ed. (Nashville: Broadman Press, 1934), 759-61, 791; Georg Benedikt Winer, A Grammar of the Idiom of the New Testament: Prepared as a Solid Basis for the Interpretation of the New Testament, rev. and ed. by Gottlieb Lünemann, trans. Joseph Henry Thayer (Andover: W. F. Draper, 1869), 112-14; Georg Benedikt Winer, A Treatise on the Grammar of New Testament Greek: Regarded as a Sure Basis for New Testament Exegesis, trans. W. F. Moulton, 3d ed. (Edinburgh: T. & T. Clark, 1882), 139-40.
6 See definition #7, #8 and #10 for γίνομαι in Walter Bauer, A Greek-English Lexicon of the New Testament and Other Early Christian Literature, rev. and ed. by Frederick William Danker, 3d ed. (Chicago: University of Chicago Press, 2000), 199. See definition #2 for ὑπάρχω, also in Bauer, Lexicon, 1029. Cf. William Watson Goodwin, Greek Grammar, rev. by Charles Burton Gulick (Boston: Ginn and Co., 1930), 198; James Hope Moulton, An Introduction to the Study of New Testament Greek, 2d ed. (London: Charles H. Kelly, 1903), 168; Robertson, Grammar, 394.
7 Nelson, “Predicate Nominative,” 14-15. Perhaps this is another way to distinguish between functional (copulative) and lexical (predicative) characteristics.
8 The first query searches in the LXX1, LXX2, NT and AF of Accordance 5.1 looked for SPN constructions involving any of the three verbs. For γίνομαι, the ANNPVG search construct yielded 64 verses (LXX1=30, LXX2=4, NT=24, AF=36). None qualified as an SPN construction. The ANVGNP search construct yielded 87 verses (LXX1=47, LXX2=4, NT=27, AF=9). None qualified as an SPN construction. The NPVGAN search construct yielded 89 verses (LXX1=47, LXX2=5, NT=30, AF=7). None qualified as an SPN construction. The VGANNP search construct yielded 86 verses (LXX1=68, LXX2=3, NT=11, AF=4). None qualified as an SPN construction. The VGNPAN yielded 107 verses (LXX1=86, LXX2=1, NT=16, AF=4). None qualified as an SPN construction. For ὑπάρχω, the ANNPVU search construct yielded 5 verses (LXX1=3, LXX2=2, NT=0, AF=0). None qualified as an SPN construction. The ANVUNP search construct yielded 4 verses (LXX1=4, LXX2=0, NT=0, AF=0). None qualified as an SPN construction. The NPVUAN search construct yielded 0 verses (LXX1=0, LXX2=0, NT=0, AF=0). None qualified as an SPN construction. The VUANNP search construct yielded 0 verses (LXX1=0, LXX2=0, NT=0, AF=0). None qualified as an SPN construction. The VUNPAN yielded 1 verses (LXX1=1, LXX2=0, NT=0, AF=0). None qualified as an SPN construction. By way of contrast, 14 examples were found for εἰμί. The ANNPVE search construct yielded 377 verses with potential matches (LXX1=231, LXX2=17, NT=93, AF=36). One from the NT qualifies as an SPN construction. The ANVENP search construct yielded 358 verses (LXX1=264, LXX2=12, NT=82, AF=28). One from the AF qualifies as an SPN construction. The NPVEAN search construct yielded 396 verses (LXX1=250, LXX2=14, NT=101, AF=31). A total of ten qualified - five from the NT, four from the LXX1, and one from the LXX2. The VEANNP search construct yielded 361 verses (LXX1=229, LXX2=10, NT=95, AF=27). One from the LXX1 qualified. The VENPAN yielded 390 verses (LXX1=239, LXX2=18, NT=106, AF=27). One from the LXX1 qualified.
9 Winer, Grammar, 521. Cf. Buttmann, Grammar, 136-37; and Goodwin, Grammar, §879.
10 The most common interpretive problem involves deciding whether the verbless cluster is a SPN construction or if the two nominatives are in simple apposition.
11 A. T. Robertson and others have observed that in the majority of SPN constructions, the predicate noun was anarthous. However, this statistic holds true largely due to the inclusion of verbless constructions in their studies, an observation confirmed by Edwin A. Blum, “Studies in the Problem Areas of the Greek Article” (Th.M. thesis, Dallas Theological Seminary, 1961), 21.
12 Wallace, Greek Grammar, 2.
13 Dana and Mantey, Manual Grammar, 68-69.
14 Ibid., 69.
15 Some additional research is needed in this area. For example, identifying all the unambiguous examples of simple apposition for the nominative, genitive, and accusative case will show patterns of proximity in these constructs. I speculate that the speaker’s more probable choice for expressing “assumption” rather than “assertion” between two nouns is to place them as close as possible to one another. In other words, the subject in an SPN construction will be both first in order (focus, emphasis) and not too far removed from the verb (greater “assumption”).
16 Wallace, Greek Grammar, 2.
17 Ibid., 43, 45.
18 Luke 1:63 reads, “His name is John.” And Matthew 13:55 reads, “Is not this the carpenter’s son? Is not his mother called Mary . . .” As previously stated, my research captured every occurrence of the articular ὄνομα and raises some questions regarding the claim that it is an exception to Wallace’s proper name rule. Regardless, the Luke passage is used here because it makes the point quite well.
19 In addition to these see Josephus, Antiquitates Judaicae 1.248.3. The passage describes how Abraham’s servant came to know the identity of the maiden for whom he was sent. The semantic subject from the context of 1.242-249 is the discovering of Rebecca’s lineage. In other words, the question lingering in the mind of the reader after a maidservant offers the servant water (and after she reveals her name, ἀλλὰ καὶ τὸ γένος ἀπεσήμαινε καὶ “῾Ρεβέκκα μέν‚”φησίν‚“ἐγὼ καλοῦμαι, . . .”) is “is she ‘the’ Rebecca?” At this point Josephus’ audience asks, “Is she related to Abraham? Who is her father?” The rest of her words provide the answer, πατὴρ δέ μοι Βαθουῆλος ἦν:ἀλλ᾿ ὁ μὲν ἤδη τέθνηκε‚Λάβανος δὲ ἀδελφός ἐστιν ἡμέτερος τοῦ τε οἴκου παντὸς σὺν τῇ μητρὶ προνούμενος καὶ τῆς ἐμῆς παρθενίασ ἐπιμελόμενος. The broader context, the immediate context (i.e., Rebecca giving her name), and the possessive pronoun show that “father” is the subject. Interestingly, the first nominative is the semantic subject in every one of these examples of convertible propositions.
20 Wallace, Greek Grammar, 4.
21 Most define this as the period of 330 B.C. to A.D. 330. However, for the purposes of this investigation I have shifted it back to cover 400 B.C. to A.D. 300.
22 Wallace, Greek Grammar, 5.
23 Bauer, Lexicon; Henry George Liddell and Robert Scott, A Greek-English Lexicon, rev. by Henry Stuart Jones and Roderick McKenzie, 9th ed. (Oxford: Clarendon Press, 1996).
24 The primary reason for disqualifying substantival participles and adjectives from the target cluster is that compared to the anarthrous noun substantive they are affected rather than unaffected elements. However, Wallace’s explanation of the inherent differences may provide another reason to exclude them from the target cluster. Substantival participles “tend to focus on activities within an (often unstated) time-frame which may or may not be characteristic” and substantival adjectives “tend to focus on character, while nouns tend to accent identity” (Daniel B. Wallace, “The Article with Multiple Substantives Connected by Kaí in the New Testament: Semantics and Significance” [Ph.D. diss., Dallas Theological Seminary, 1995], 142-43, 233-36). Some examples of substantival adjectives and participles are provided nonetheless in table two in appendix four for the purpose of corroboration.
25 Robert Young, Analytical Concordance to the Bible, 22d American ed. (Grand Rapids: Eerdmans, 1975), 1-23 (found as repagination at the end of the book). The characteristic of nonpluralization is a generally accepted criterion and sufficient identifier for whether a noun is proper but Wallace insightfully comments, “[T]he necessity of nonpluralization needs to be nuanced. On the one hand, simply because a particular word does not occur in the plural in the extant literature is no guarantee that it is a proper name . . . . On the other hand, it is possible on a rare occasion for even a proper name to be pluralized” (Wallace, “Article,” 164, see esp. his fn. 207). However, he rightly points out that these occurrences are rare and that “it may be taken as a principle of Greek grammar that proper names do not pluralize” (Wallace, “Article,” 165).
26 Wallace provides a very helpful discussion on the major difference between place names and proper names. Essentially, place names have the characteristic of “referential overlap” while proper names do not (Wallace, “Article,” 166). For example, Mt. Whitney and California both incorporate Mt. Whitney in the reference. However, the semantic difference between place names and proper names does not significantly affect the aim of my investigation primarily because, at the unaffected level, place names denote more than they connote; they have “reference but not sense” (Wallace, “Article,” 85, 100, 165).
27 Philip B. Harner, “Qualitative Anarthrous Predicate Nouns: Mark 15:39 and John 1:1,” Journal of Biblical Literature 92 (1973); Paul Stephen Dixon, “The Significance of the Anarthrous Predicate Nominative in John” (Th.M. thesis, Dallas Theological Seminary, 1975); Wallace, “Article.”. The classic and much debated example is in John 1:1c, which Wallace persuasively argues to be qualitative (Wallace, Greek Grammar, 266-69). Much of this work can be more fully appreciated by consulting his doctoral dissertation (Wallace, “Article,” 101, 164-65, 260-63).
28 The noun appears quite a bit through out the NT (χριστός appears 529 times by itself and 107 times with ᾿Ιησοῦς).
29 Bauer, Lexicon, 1091. Cf. Wallace, “Article,” 164, 165, 261 fn. 24.
30 This tentativeness may be philosophically analogous to Wallace’s approach to discussing the validity of Granville Sharp’s rule regarding article-substantive-καί-substantive constructions, where Wallace first examines data which do not contain christologically significant texts. See Wallace, Greek Grammar, 273-77.
31 Levinsohn, Discourse Features, 100.
32 Cf. Barnabas 8:2 and Ignatius: To the Philadelphians 8:2
33 One lexicon labels Ναζαρηνός as an adjective. See Bauer, Lexicon, 664. For similar constructions see John 5:15, 11:2, and 18:14.