Smart's Rule: A Critique

Robert M. Bowman, Jr.

Since 2000, a particular claim concerning biblical Greek grammar known as �Smart�s rule� has been circulating on the Web. This rule is thought by its advocates to prove that in John 20:28 Thomas must have been speaking of two persons and not one when he said �My Lord and my God!� That is, the rule supposedly proves that Thomas was speaking of Jesus as his Lord but of the Father as his God.

Oddly enough, I cannot find any exposition of the rule from Smart himself. Mr. Smart appears to be Martin Smart, who was and, I presume, still is a Jehovah�s Witness. Beyond that I have no information about Mr. Smart and have not seen anything in writing from him presenting or defending his own rule.

Most of the references to Smart�s rule on the Web appear on a listserv known as B-Greek, in its archived discussions from late 2000 and early 2001 (http://www.ibiblio.org/bgreek/). Recently, on a listserv that I moderate (http://groups.yahoo.com/group/evangelicals_and_jws/), one of the list members touted Smart�s rule as superior to Sharp�s rule, which is a well-known and much-discussed rule of Greek grammar.

Let us, then, take a look at Smart�s rule and see if there is anything to it.

Defining Smart�s Rule

The writer who actually presented and defended Smart�s rule on B-Greek was a Jehovah�s Witness named Dan Parker. Here is how he defined the rule:

�In native [not translation] KOINE Greek when the copulative KAI connects two substantives of personal description in regimen [i.e. both or neither have articles] and the first substantive alone is modified by the personal pronoun in the genitive or the personal pronoun is repeated for perspicuity [Winer 147-148;155] two persons or groups of persons are in view.�  Dan Parker, �Re: John 20:28 and Smart�s rule. Correction,� B-Greek listserv, 2/1/2001 ;

http://www.ibiblio.org/bgreek/test-archives/html4/2001-02/4231.html (bracketed material in original).

Note that Smart�s rule as given by Parker actually covers four types of constructions:

Article + Substantive + Pronoun + kai + Article + Substantive + Pronoun

Substantive + Pronoun + kai + Article + Substantive + Pronoun

Article + Substantive + Pronoun + kai + Article + Substantive

Substantive + Pronoun + kai + Article + Substantive

It does not matter whether the pronoun precedes or follows the substantive.

The way that Smart�s rule is worded, it may seem to be qualified in such a way that it need not apply to all texts using the third and fourth constructions (i.e., those that do not repeat the pronoun �for perspicuity�). However, from Parker�s contention that the rule proves that John 20:28 is speaking of two persons, I infer that what he (and presumably Smart) mean is that whenever the pronoun is repeated with the second substantive the repetition is for the sake of perspicuity and that in all such cases the second noun still refers to someone different than the first substantive.

According to Parker, then, the rule has no exceptions in biblical Greek when the following conditions are met:

1. The text is not translation Greek (i.e., the Greek OT is entirely excluded).
2. There are two substantives of personal description joined by KAI.
3. Either both substantives have an article or neither of them has one.
4. Either the first substantive alone, or both substantives, are modified by a personal pronoun in the genitive (i.e., texts in which neither is so modified, or in which the second substantive alone is so modified, are not included).

The Repetition of the Pronoun

As already noted, in constructions thought to be governed by Smart�s rule a pronoun in the genitive case must be attached to the first substantive and may be attached to the second substantive �for perspicuity.� I am not sure why this qualification is attached since Parker and others who endorse the rule seem to think that it applies to any text in which the pronoun is repeated with the second substantive.

As stated, the rule refers to a few pages in Winer to document this idea of the repetition of the pronoun for perspicuity. The fact is that Winer was not referring to a doubling of the same pronoun or even necessarily the use of two pronouns. Rather, he was speaking of the use of a pronoun that in some sense is redundant but is used because the antecedent is already several words or more distant:

4. A repetition of this pronoun (autos), and also of the other personal pronouns, occurs,

a. When subjoined for the sake of perspicuity, in sentences where the principal noun is followed by a number of other words�. In the majority of these passages a participial construction, equivalent to an independent clause, precedes; in this same case even the Greek authors often add the pronoun [citing Pausanius, Herodotus, Plato, et. al.]�.

[G. B. Winer, A Grammar of the Idiom of the New Testament prepared as a solid basis for the interpretation of the New Testament, 7th ed., enl. and improved by Gottlieb L�nemann, rev. and authorized translation (Andover: W. F. Draper, 1897), 147-48.]

The NT texts that Winer cites as examples of his point often do not even use two pronouns. Instead, they use a pronoun to refer back to an antecedent substantive that is far enough back in a somewhat complex sentence structure, so that the pronoun helps to make the antecedent clearer:

�The people who sat in darkness have seen a great light, and to those who were sitting in the land and shadow of death, a light has dawned to them [autois]� (Matt. 4:16).

�Now when he had gone out to the entrance, another woman saw him [auton]�� (Matt. 26:71).

�Therefore, the one who knows the good thing to do and does not do [it], to him [aut�] it is sin� (James 4:17 ).

In Winer�s other examples, sometimes two pronouns are used and they are the same (Acts 7:21 , auton�auton; Col. 2:13, humas�humas; Rev. 6:4, aut�aut�), sometimes the two pronouns are of different cases (Mark 5:2, autou�aut� [Winer has aut�aut�]; Mark 9:28 , autou�auton; Matt. 8:1, autou�aut�), and in a couple of instances it is hard to tell which pronoun Winer meant (Matt. 5:40; Phil. 1:7). None of Winer�s examples are examples of what is called Smart�s rule.

As best I can tell, then, Winer�s treatment of the pronoun used for perspicuity has no bearing or relevance to Smart�s rule.

Alleged Examples of Smart�s Rule

In one of his posts to B-Greek, Dan Parker offered the following lists of examples of texts fitting Smart�s rule. This same list has appeared verbatim elsewhere, notably on a web site defending Jehovah�s Witness doctrine (�An Online Response to a Kevin Quick Defender."  The website where this article originally appeared has been taken down.  As of Aug. 2005, it is available for purchase at the following site:  http://www.geocities.com/yhwhbible/main.htm)

Possessive pronoun repeated for perspicuity (21) - (Mt 12:47,49; Mk 3:31 ,32 ,33 ,34 ; 6:4 7:10 ; 8:20, 21 Lu 8:21; Jn 2:12; 4:12; Ac 2:17; Ro 16:21 ; 1Th 3:11 ; 2Th 2:16 ; 1Ti 1:1; 2Ti 1:5; Heb 8:11; Re 6:11) [Heb 1:7 is a LXX quote and is therefore translation Greek.]

Single possessive - both substantives anarthrous (10) - (Mk 3:35; Ro 1:7; 1Co 1:3; 2Co 1:2; Ga 1:3; Ep 1:2; Php 1:2; 2Th 1:1,2; Phil 1:3)

Single possessive pronoun - both substantives arthrous (12) - (Mk 6:21; 10:7,19; 16:7; Lk 2:23; 14:26; 18:20; Jn 11:5; Eph 6:2; Ac 7:14; 10:24; Re 11:18)

This is the only list I have been able to find of alleged examples.

Before going any further, some corrections to the list of verse references are needed. The first paragraph appears to cite Mark 8:20, 21, followed by Luke 8:21, but in fact the first two references should be to Luke 8:20, 21 (so that the next one is actually a repeat). There are thus only 20 verses listed in the first paragraph. In the second paragraph, �Phil 1:3� is a reference to Philemon 3. In the third paragraph, the first reference in Luke should be to Luke 2:33, not 2:23 .

Next, in the interest of giving Smart�s rule every chance, we should take note of other texts that appear to fit its parameters successfully. I have found a few other such references: Matthew 12:50; 13:55 ; and John 19:25.

We have, then, 42 references listed by Dan Parker (not 43, since he accidentally counted Luke 8:21 twice), plus three more references, for a total of 45 NT references (to my knowledge) that have been or might be cited as examples of Smart�s rule.

The Semantics of Conjoined Substantives

The first question that needs to be asked about these example texts is whether there would be any possibility of the substantives being understood as referring to a single referent regardless of the articles or pronouns attached to them. The answer in most of these texts is an unequivocal no. The semantic relation between the substantives alone is enough to make their different referents unambiguous.

In 26 of our 45 examples, the substantives include terms designating family relationships that must be held by different individuals. These include:

�      6 texts that speak of a person�s father and mother (Mark 7:10 ; 10:7, 19; Luke 2:33 ; 18:20 ; Eph. 6:2)

�      12 texts that speak of a person�s mother, brothers, and sisters (Matt. 12:47 , 49, 50; 13:55 ; Mark 3:31 , 32, 33, 34, 35; Luke 8:20 , 21; John 19:25 ).

In addition, the texts include references to the following:

�      sons and daughters (Acts 2:17 )

�      Timothy�s grandmother and mother (2 Tim. 1:5)

�      one�s father, mother, wife, children, brothers, and sisters (Luke 14:26 )

�      Jesus (referred to as �he�) and his mother, brothers, and disciples (John 2:12 )

�      Jacob (�himself�), his sons, and his cattle (!) (John 4:12)

�      Martha, her sister, and Lazarus (John 11:5)

�      Jacob his (Joseph�s) father and all his relatives (Acts 7:14 )

�      Cornelius�s relatives and friends (Acts 10:24 )

Since it is impossible for any of these texts to be referring to a single individual, or even to be using the conjoined plural nouns to refer to the same group (since one�s sons cannot also be one�s daughters or one�s cattle, for example!), these texts cannot tell us anything as to the possible semantic significance of the placement of the possessive pronouns with regard to the nouns not having the same referent.

Twelve of the remaining examples refer to God the Father and to Jesus Christ (Rom. 1:7; 1 Cor. 1:3; 2 Cor. 1:2; Gal. 1:3; Eph. 1:2; Phil. 1:2; 1 Thess. 3:11; 2 Thess. 1:1, 2; 2:16; 1 Tim. 1:1; Phm. 3). All but two of these (1 Thess. 3:11 ; 2 Thess. 2:16 ) occur in a Pauline salutation, and all but one (1 Tim. 1:1, which has �God our Savior and Christ Jesus our hope�) uses essentially the same wording: �God our Father and [the] Lord Jesus Christ� (note the slight variations in the two non-salutation examples). In all 10 of the salutation examples, the two compound noun groups are anarthrous. The style and setting in Paul�s letters in all of these 10 examples fits the usual Greek opening of a letter at that time, in which good wishes were passed along from the letter writer�s companions closest to the recipients (�Ted to Mary. Cheers from Dad and Bill. We wish you were here.�) Even if we didn�t already know that God our Father was someone different from the Lord Jesus Christ, we would know simply by reading these salutations in their context that the two were different referents. The use or nonuse of the pronoun �our� to modify �Father� would make absolutely no difference in how these salutations would be read.

This leaves only seven examples of constructions supposedly governed by Smart�s rule. That isn�t much on which to base a rule of grammar. But let�s look at these remaining texts to see what evidence they might yield in support of Smart�s rule.

Weak Support for the Rule

In Mark 6:4, Jesus says, �A prophet is not without honor except in his hometown and among his relatives and in his household.� There are two reasons for questioning whether this text fits the parameters of Smart�s rule at all. First, the terms �hometown� (patris) and �household� (oikia) do not describe persons in the way that �relative� (sungen�s) does. Rather, they refer to entities composed of persons. I am unsure whether Smart�s rule could be extended to include such substantives, assuming the rule were valid. Second and more telling, the three nouns are not simply connected by kai, as Smart�s rule assumes, but are parts of three prepositional phrases using en (�in,� �among�) connected by kai. That having been said, in this instance we do not have terms referring to separate referents. Jesus� statement proceeds from the largest unit (his hometown) to a subset of that unit (his relatives) to a subset of that unit (his own family or household).

Mark 6:21 speaks of Herod�s �leaders and the commanders and the leading men of Galilee .� These do appear to be three distinct groups of men, and so this text apparently does fit Smart�s rule.

In Mark 16:7, the angel at the tomb tells the women to tell �his disciples and Peter.� Of course, Peter was one of Jesus� disciples, so here again, as in Mark 6:4, the second term is a subset of the first.

In Romans 16:21, Paul writes, �Timothy my fellow worker greets you, and also Lucius and Jason and Sosipater, my kinsmen.� Grammatically, Paul connects the four proper names together with kai; the expression �my fellow worker� is in apposition to �Timothy� and the expression �my kinsmen� is in apposition to the other three names. This text, then, is a dubious example of Smart�s rule.

Hebrews 8:11 is a quotation from the Greek Old Testament, and so according to Parker�s definition cannot be used in support of Smart�s rule. That is probably just as well, since the substantives are arguably synonyms: �And they shall not teach everyone his fellow-citizen [polit�n] and everyone his brother [adelphon]�.� In this context �brother� probably means something like �fellow countryman.� At any rate, the two nouns do not refer to separate groups.

Revelation 6:11 refers to �their fellow servants and their brothers who were about to be killed just as they had been.� In this text the martyred believers in Christ are told to rest while others are martyred for their faith just as they had been. These others are described as �their fellow servants [sundouloi] and their brothers [adelphoi].� Since all believers are considered servants of Christ and brothers of one another, the two nouns here refer to the same group. Therefore, this text appears to be a fairly clear counterexample to Smart�s rule. Furthermore, it is a text in which the possessive pronoun is repeated, as in John 20:28.

Finally, Revelation 11:18 the time is said to have come for God to reward �your slaves the prophets and the saints and those who fear your name� (tois doulois sou tois proph�tais kai tois hagiois kai tois phoboumenois to onoma sou). God�s servants the prophets are presumably a subset of the saints, and indeed of those who fear God�s name. It is also likely that the saints are identical with those who fear God�s name. This text, therefore, does not seem to fit Smart�s rule very well, and may be at least partially contrary to Smart�s rule.

Of these seven texts, only one clearly fits what Smart�s rule claims (that the two substantives have different referents), namely, Mark 6:21 . Romans 16:21 probably is irrelevant because the two substantives qualified by possessive pronouns are in apposition to proper names and are not strictly speaking in regimen with each other. In all of the other texts, the two or three substantives are either synonymous or one substantive is a subset of the other. Revelation 6:11 is a reasonable clear counterexample to Smart�s rule.

More Counterexamples to Smart�s Rule

So far I have found two more clear counterexamples to Smart�s rule.

2 Corinthians 8:23. �As for Titus, he is my partner and fellow worker among you [koin�nos emos kai eis humas sunergos]�.� Note that the nouns �partner� (koin�nos) and �fellow worker� (sunergos) both explicitly describe the same person, namely, Titus. This text, then, is a clear counterexample to Smart�s rule.

Philippians 2:25. �But I thought it necessary to send to you Epaphroditus, my brother and fellow worker and fellow soldier (ton adelphon kai sunergon kai sustrati�t�n mou), who is also your messenger and minister to my need (hum�n de apostolon kai leitourgon t�s xreias mou).� Remarkably, in this one verse we have an illustration of Sharp�s rule in the first half of the verse and a clear counterexample of Smart�s rule in the second half of the verse. The noun string ton adelphon kai sunergon kai sustrati�t�n perfectly fits Sharp�s rule. The second noun string describing Epaphroditus, hum�n de apostolon kai leitourgon t�s xreias mou, fits the syntax structure of Smart�s rule (possessive pronoun�noun�kai�noun, with the two anarthrous nouns in regimen), yet the nouns both have the same referent. Therefore, this text also is a clear counterexample to Smart�s rule.

Conclusion

Smart�s rule finds all of its support in texts where the two or three nouns in regimen must have different referents because of the semantics of the nouns in question (e.g., father and mother, or mother and sons). In texts where this is not the case, more often than not the nouns do not have separate referents. Sometimes the referents are overlapping, or one refers to a subset of the other noun; and sometimes the referent is identical.

The bottom line is that the presence or absence of a possessive pronoun has no bearing on the exegesis of phrases in which two or more substantives are connected by kai. Where the substantives are in regimen (all having the article or all lacking the article), the meaning of the terms in context is the only real consideration in determining whether the substantives have one or more referent. In texts where the substantives used are generally synonyms, as in John 20:28 (�Lord� and �God�), the presumption should be that they have the same referent unless the context explicitly indicates otherwise.