Biblical topics
Bible Study
Pope Shenouda
Father Matta
Bishop Mattaous
Bishop Moussa
Bishop Alexander
Habib Gerguis
Fasts & Feasts
Family & Youth
3ds Max 2016
Account Payable
Accounts Receivable
Active Directory
Adaptive Access Manager
Adobe Premiere Pro
Apache Hive
Asset Management
Big data
Building OA Framework
Business Intelligence
C Sharp
Cash Management
Crystal Reports
Data Acquisition
Data Architecture
Data Archiving
Data Guard
Data Mining
Data Modeling
Data Structure
Data Visualization
Design Illustration
Expression Web
Fusion Middleware
General Ledger
Google Drive
GoPro Studio
Hyperion Planning
Massive UE4
MS Access 2016
MS Exchange Server
MS OneNote 2016
MS OneNote 2016 
MS Outlook 2016
MS PowerPoint 2016
MS Publisher 2016
MS SharePoint 2016
MS Word
Oracle 12c Administration
Oracle EBS
Oracle E-business tax
Oracle Financial Applications
Oracle Identity Manager
Oracle Mobile
Oracle Payroll Fundamentals
Oracle Performance Tuning
Oracle Product Lifecycle
Oracle project
Oracle Purchasing
Oracle RAC admin
Oracle SOA admin
Photoshop CS6
Project Management
R Programming
SQL Server
Subledger Accounting
Supply Chain Planning
Visual Basic
Visual Studio
Weblogic Server
Windows 10
Windows Server




                   NEW TESTAMENT













                                             William E. Elliott














                        Submitted in partial fulfillment of requirements

                              for the degree of Doctor of Theology in

                                        Grace Theological Seminary

                                                      May 1981


                      Digitized by Ted Hildebrandt, Gordon College 2006



Author:         William E. Elliott

Degree:        Doctor of Theology

Date:            May 1981

Committee: Dr. Charles R. Smith, Dr. John A. Sproule, Dr. Homer A. Kent, Jr.


          Increasing interest in the grammar of the Greek New Testament

has focused attention upon aspects of the language that have, for the

most part, been passed over by past grammarians. Among these topics

is that of conditional sentences. A superficial survey of the lit-

erature indicates that most writers seem to have the data confidently

in tow, but closer inspection shows that this is not the case. Modern

grammarians are, for the most part, content to follow the lead of A.T.

Robertson and classify these clauses in terms of First, Second, Third,

and Fourth Class conditions. Others, dissatisfied with Robertson's

system and the extension of it by his followers, have returned to the

terminology, if not the principles of Classical Greek. The situation

is uncoordinated at best, for even in Classical Greek studies there

is significant disagreement upon the classification of these sentences.

          The historical background to the study of conditional sentences

is presented from both the Classical and the Koine Greek standpoints.

Suggested systems of classification include Time, Fulfillment, Form

and Determination. The latter, championed by B.L. Gildersleeve, is

the preferred system. Determination is indicated by the mood of the

verbs employed in the protasis. Gildersleeve's system entered Koine

studies primarily through the work of A.T. Robertson. He employes

four classes into which he places these conditional sentences.

          The Simple Condition, using the indicative mood, states the

condition as an assumed reality. There is no necessary connection

between actuality and the statement. This condition merely presents

the conclusion as a necessary corollary of the condition.

          The Contrary to Fact Condition also uses the indicative mood

to present the condition as one that is assumed not true, i.e., con-

trary to fact. Again, there is no necessary connection between

actuality and the conditional statement.

          The Probable Condition presents the condition as one assumed

probable, i.e., one that could easily be fulfilled. The hypothetical

nature of this condition requires the use of the subjunctive mood.

          The Possible condition states the condition as one that is

assumed possible, i.e., little likelihood of fulfillment. This con-

dition utilizes the optative mood, and there is no complete example

of it in the New Testament.

          Two basic concepts underlie all conditional sentences. First,

the determining factor is the mood of the verb, not the particle em-

ployed. Second, all conditional sentences state their case as an

assumption, never as a direct statement of reality.
















            Accepted by the Faculty of Grace Theological Seminary

                in partial fulfillment of requirements for the degree

                                          Doctor of Theology



                                         Examining Committee:


                                        Charles R. Smith


                                                 Homer A. Kent Jr.


                                         John A. Sproule



                             TABLE OF CONTENTS


INTRODUCTION                                                                                 1



I.   A HISTORICAL SURVEY                                                               4

          Conditional Sentences in General                                                  4

          Conditional Sentences in Classical Greek                                                8

                    The Importance of Classical Greek                                      10

                    Suggested Classification Systems                                                 10

                              Classification According to Time                              11

                              Classification According to Fulfillment                     18

                              Classification According to Form                              20

                              Classification According to Determination                 23

          Conditional Sentences in Koine Greek                                           33

                    Early Grammarians                                                                       34

                              George Benedict Winer                                             34

                              Alexander Buttmann                                                37

                              Samuel G. Green                                                      40

                              Ernest DeWitt Burton                                               42

                              James Hope Moulton                                               46

                    Modern Grammarians                                                        49

                              Archibald Thomas Robertson                                   49

                              William Douglas Chamberlain                                   55

                              Charles Francis Digby Moule                                    57

                              Friedrich Blass - Albert Debrunner                            60

                              Nigel Turner                                                            62

                              H. E. Dana and Julius R. Mantey                                         65

                              Robert W. Funk                                                       67

                              William Sanford La Sor                                            65

                    Summary of Koine Grammarians                                        71


II.   THE SIMPLE CONDITION                                                              73


          Introduction                                                                                 73

          The Conditional Particle ei                                                             75

                    The Significance of ei                                                  75

                    The Significance of ei with other Particles                            77

          The Negative Particles in Simple Conditions                                  81

          Ean with the Indicative Mood                                                        84

          Significance of Moods and Tenses                                                          88




          Meaning of the Simple Condition                                                  95

                    Particular and General Conditions                                                 95

                    Degree of Reality                                                               98

          Translation of the Simple Condition                                               105


III.      THE CONTRARY TO FACT CONDITION                                  106

          Introduction                                                                                 106

          Significance of Tense                                                                   110

                    Imperfect Tense                                                                 110

                    Aorist Tense                                                                                 113

                    Pluperfect Tense                                                                 115

          The Use of An in Contrary to Fact Conditions                                 116

          The Meaning of the Contrary to Fact Condition                                       121

          The Translation of the Contrary to Fact Condition                          123


IV.     THE PROBABLE CONDITION                                                   129

          Introduction                                                                                 129

          Analysis of the Probable Condition                                                131

                    The Protasis                                                                       131

                    The Apodosis                                                                     141

          Meaning of the Probable Condition                                               148

                    Review of the Grammarians                                               148

                    Evaluation of the Grammarians                                           151

                    Relationship with the Simple Condition                               156

                    Summary                                                                           163

          Translation of the Probable Condition                                           165


V.      THE POSSIBLE CONDITION                                                      169

          Introduction                                                                                 169

          The Optative Mood in General                                                      170

          The Optative Mood in Conditional Sentences                                 174

                    The Construction                                                                175

                    The Significance                                                                 178

                    The Grammarians                                                              178

                    The Specific Examples                                                       180

                              Those with the Protasis Implied                                180

                              Those with the Protasis Stated                                   184

          Translation of the Possible Condition                                            191


VI.     CONCLUSION                                                                            193

          Simple Conditions                                                                        194

          Unreal Conditions                                                                        195

          Probable Conditions                                                                               195

          Possible Conditions                                                                      196

          Summary                                                                                     196




APPENDICES                                                                                       198

          Appendix I: Occurrences of the Simple Condition                          198

          Appendix II: Occurrences of the Unreal Condition                         211

          Appendix III: Occurrences of the Probable Condition                     215


BIBLIOGRAPHY                                                                                   226








          God created man with the potential for abstract reasoning, and

his many languages reflect this through their use of the subjunctive

mood: the mood of contingency or possibility. Posing questions,

exploring possibilities and analyzing logical connections are part of

man's reasoning capabilities, and his languages reflect these skills.

Among the syntactical tools which accomplish these are conditional

sentences. These sentences, usually consisting of two clauses,

state a hypothesis and give a conclusion. In English this corres-

ponds to the "If . . . then" formula.

          This type of sentence, while prevalent in English, is usually

listed as one of several subordinate clause relationships.1  By

contrast, the Greek language presents a more fully developed system

of conditional sentences by means of which a remarkable degree of

precision may be obtained in expressing conditional thought. The

Greek conditional sentence presents both the condition and certain

specific implications about it in one sentence whereas English needs

both the conditional statement and qualifying sentences to communicate

the same concept. This compactness lends itself to greater precision


          1 Porter Perrin and George H. Smith, Handbook of Current

English, third edition, edited by Jim W. Corder (Glenview, Illinois:

Scott, Foresman and Company, 1968), 48-56, 120-121.



in the statement of conditional concepts.

          Since conditional sentences are basic to the material of the

Greek New Testament, a detailed understanding of conditional sentences

is vital for an accurate interpretation of its contents. This, then,

is the goal of this study: to explore conditional sentences so that the

message of the New Testament may be better understood.

          Though all students of Greek, both Classical and Koine, agree

on the importance of conditional sentences, few agree on the analysis

of them. A. T. Robertson aptly describes the situation in Koine studies

when he writes, "In truth the doctors have disagreed themselves and the

rest have not known how to go."1 The Classical scene is likewise

muddled, as Blass-Debrunner notes, "The classical grammars are also hope-

lessly at variance."2

          Some of this confusion is due to the absence of a standard by

which to classify conditional sentences. Time, degree of reality and

construction have all been suggested by various grammarians as possible

classification systems. Further, each grammarian seems to have developed

his own terminology in discussing the subject, and each argues that his

is best. Indeed, it is possible to trace the influence of major

grammarians through succeeding generations by noting who adopts their

terminology in dealing with conditional sentences.


          1 A. T. Robertson, A Grammar of the Greek New Testament in the

Light of Historical Research (Nashville, Tennessee: Broadman Press,

1934), p. 1004.

          2 Friedrich Blass and Alvert Debrunner, A Greek Grammar of the

New Testament and Other Early Christian Literature, translated and

revised by Robert W. Funk (Chicago: The University of Chicago Press,

1961), p. 189.



          Another reason for the lack of standardization may be the

inherent flexibility of the language itself. Though Classical and Koine

Greek may be considered fossilized ancestors of Modern Greek, they were

living, functioning languages, complete with the internal syntactical

elasticity found in living languages today. Greek, like English, developed

through usage, and patterns so developed may refuse to be forced into a

logically consistent mold. So, whether through lack of a standard, or

lack of accepted terminology, or through syntactical flexibility,

conditional sentences have provided grammarians with a fruitful area of

contemplation, and students with a frustrating area of concentration.

          This study seeks to offer help to those involved in the

analysis of conditional sentences by summarizing the work of previous

grammarians and giving a detailed analysis of each type of conditional

sentence in the New Testament. The work of past and contemporary

scholars will be surveyed to give an overview of their studies, agree-

ments, and disagreements. Then the conditional sentences in the Greek

New Testament will be identified and analyzed with the help of principles

obtained from the grammatical survey. Finally, observations will be

offered on the important matters of translation and interpretation.

The result should be a small but positive step in gaining further insight

into the meaning of conditional sentences in the Greek New Testament.





                                      CHAPTER I


                           A HISTORICAL SURVEY


          Since any study necessarily builds upon the work of others, a

survey of previous studies of conditional sentences is basic to a

thorough understanding of the topic. This study will include the work

of both Classical and Koine scholars.

                         Conditional Sentences in General

          A brief survey of the technical details of conditional sentences

will set the scene for the succeeding discussion and evaluation.


                      The Definition of Conditional Sentences

          A conditional sentence is a two-clause sentence in which the

first clause states a supposition or hypothesis and the second clause

states the results if that condition is met. The hypothetical clause

which states the condition ("If this . . .") is termed the protasis

and the conclusion clause is called the apodosis (". . . then this.").

Herbert Weir Smyth explains it this way:


          A condition is a supposition on which a statement is based.

A conditional sentence commonly consists of two clauses:

          The protasis: the conditional, or subordinate, clause,

expressing a supposed or assumed case (if).

          The apodosis: the conclusion, or principal, clause, expressing

what follows if the condition is realized. The truth or fulfillment

of the conclusion depends on the truth or fulfillment of the

conditional clause.1


          1 Herbert Weir Smyth, Greek Grammar for Colleges (New York:

American Book Company, 1920), p. 512.



                The Construction of Conditional Sentences

The Protasis

          As Smyth stated, the protasis stands as the subordinate or

dependent clause, setting forth the condition. The term protasis comes

from prosta<sij, "lit. stretching forward, that which is put forward

(in logic, a premise).1  While the formal sequence is the standard

"If . . . then," English, as well as Greek, varies the sequence in

usage: "You will receive the reward if you do a good job."

          The form of the protasis in Greek involves a conditional

particle (ei] or e]a<n) and a verb. The various combinations of particles

and verbs will be discussed later. Though the mood of the verb is the

key element in identifying the type of condition, certain constructions

are fairly standard. Again, these will be presented later. This

combination of particles and moods enables Greek to express conditional

thought with a compact precision lacking in English. The thought of

a few Greek words may take a few English sentences to be communicated.


The Apodosis

          The main or independent clause in a conditional sentence is

termed the apodosis. This term comes from "a]podo<sij, lit. giving back,

return; i. e. the resuming or answering clause."2  The apodosis may

employ verbs in any tense or mood, and frequently, in the Koine at


          1 Smyth, Grammar, p. 512.

          2 Ibid., p. 512.



least, uses the particle a@n with moods other than the indicative.1  Taken

together, then, the protasis and apodosis constitute a conditional



                                The Conditional Particles

          The origin of the Greek conditional particles is as obscure as

their usage is important. Goodwin, one of the leading Classical Greek

grammarians, succinctly states:

          It is impossible to discuss intelligently the origin of the

      conditional sentence until the etymology and original meaning of

      the particles ei], a@n, and ke< are determined. On these questions

      we have as yet little or no real knowledge.2

He then gives a brief summary of what is known about these particles

and concludes:

      But here we are on purely theoretical ground; and we must content

      ourselves practically with the fact, that in the earliest Greek

      known to us ei was fully established in its conditional sense,

      like our if and Latin si.3


          1 H. E. Dana and Julius R. Mantey, A Manual Grammar of the

Greek New Testament (New York: The MacMillan Company, 1966), p. 288.

          2 William Watson Goodwin, Syntax of the Mood and Tenses of the

Greek Verb (Boston: Ginn & Company, 1893), p. 142.

          3 Ibid., p. 143.




          Liddell and Scott trace the origin of ei] back to its use in

Homer where it is sometimes replaced by the Doric ai].1  It introduces

either conditional clauses or questions and is regularly used with the

indicative mood.2  Its consistent translation in conditional clauses is

"if." The relationship of this particle, the indicative mood and the

assumed reality of the condition will be discussed later.



          Ean is a combination of ei and an, according to Dana and

Mantey.3  Smyth remarks that "The etymology of e]a<n is uncertain:

either from h] + a@n or from ei] + a@n.”4  This particle introduces

conditions in the subjunctive mood, though it is not limited to this


      The difference between ei] and e]a<n has been considerably lessened

      in Hellenistic as compared with earlier Greek. We have seen that

      e]a<n can even take the indicative; while (as rarely in classical

      Greek) ei] can be found with the subjunctive.5


          1 Henry George Liddell and Robert Scott, A Greek-English Lexicon,

Vol. I, edited by Henry Stuart Jones and Roderick McKenzie (Oxford:

At the Clarendon Press, 1951), p. 480. See also William Watson

Goodwin, An Elementary Greek Grammar (Boston: Ginn Brothers, 1872),

p. 263.

          2 Dana and Mantey, Grammar, p. 246.

          3 Ibid., p. 245.

          4 Smyth, Grammar, p. 512.

          5 J. H. Moulton, A Grammar of New Testament Greek, Volume I:

Prolegomena, third edition (Edinburgh: T. & T. Clark, 1908), p. 187.



This caution should guard against absolute rules, but the general

principle is that e]a<n utilizes the subjunctive mood while ei] employs

the indicative. Again, the specific implications of this regarding the

assumed reality of the condition will be discussed later.

          These conditional particles are similar to particles of

interjection in Homeric Greek and related languages. N. D. Coleman

suggests that the conditional particles "appear to have been inter-

jections in the first place."1  J. B. Greenough tries to push the

origin of the conditional sentence and its attendant particles back into

the frontiers of the Indo-European linguistic heritage: "We are

naturally led to conclude that this [conditional] construction was in

use more or less in Indo-European times and was received by each of the

languages as a part of the original inheritance."2  Whatever the source

of these particles, conditional sentences were a vital part of the

language of both Classical and Koine Greek and play a vital role in the

Greek New Testament.


                     Conditional Sentences in Classical Greek

          Prior to the discovery of the papyri and the comprehension of

their linguistic significance, Biblical Greek was considered to be


          1 N. D. Coleman, "Some Noteworthy Uses of ei] in Hellenistic

Greek with a Note on St. Mark viii 12," The Journal of Theological

Studies, 27:1 (April, 1976), p. 159.

          2 James B. Greenough, "On Some Forms of Conditional Sentences

in Latin, Greek and Sanskrit," Transactions of the American Philo-

logical Association, 2:2 (June, 1871), p. 164.



a unique species of that language, sometimes referred to as "Holy Ghost

Greek."1  This was, of course, a false position, for the Greek of the

New Testament was simply the language of the common man as found in

"the street and market place."2  This was distinguished from Hellenistic

Greek, a direct descendant of Attic or Classical Greek. While some

New Testament writings show distinct Hellenistic influence, such as

Luke, Acts and Hebrews, others are distinctly Koine. This is not

because the writers, according to Moulton, used Greek "as foreigners,

Aramaic thought underlying Greek expression."3  Rather it is due to

the individual writers using the language closest to them, each reflecting

their own blend of Hebrew and Hellenistic cultures. As the result of

pioneering efforts by men like Adolf Deismann in analyzing the papyri,

"Biblical" Greek became identified as the language of the common man,

the Koine Greek. To be sure, it still reflected the Hebrew idiom of

the authors, but it was Koine none the less. A. T. Robertson sums up

the current understanding of New Testament Greek:

          The Greek of the New Testament that was used with practical

      uniformity over most of the Roman world is called the Common Greek

      or koinh<. Not that it was not good Greek, but rather the Greek

      in common use. There was indeed a literary koinh< [Hellenistic

      Greek] and a vernacular koinh<.  Plutarch is a good specimen of the,

      literary koinh< while the papyri are chiefly in the vernacular koinh<  

      like most of the New Testament.4


          1 Dana and Mantey, Grammar, pp. 9-15.

          2 W. White, Jr., "Greek Language," The Zondervan Pictorial  

Encyclopedia of the Bible, 5 vols., Merrill C. Tenney, editor (Grand

Rapids: Zondervan Publishing House, 1977), II, pp. 827-828.

          3 Moulton, Prolegomena, p. 3.

          4 A. T. Robertson, A Short Grammar of the Greek New Testament

(New York: A. C. Armstrong E. Son, 1908), p. 6.



The Importance of Classical Greek

          If the Koine of the New Testament is in the mainstream of

linguistic development and expression, does it have any significant

connection with the literary history of the language as a whole, and

with Classical Greek specifically? Yes, answers Robertson:

          This koinh< was itself the heir of the past. The various Greek

       dialects blended on an Attic base. The koinh< was thus richer in

       expression as to words and forms than any of the older dialects.

      Compare the relation of modern English to the various tongues that

      have contributed to its power and expansion. Ionic, Doric, Aeolic,

      North West Greek and other dialects have made some contribution to

      the common result. The use of nominatives in the midst of accusa-

      tives in the Boeotian, for instance, is strangely like the Book of

      Revelation. So the absence of the future participle is like the N.T.1

          This heritage, then, is sufficient justification for beginning

the study of conditional sentences in Classical Greek, the language

of the period from Homer to the Alexandrian conquests (c. 330 B.C.).2

This language constituted the "chief basis of New Testament Greek,"3

thus its handling of conditional sentences has important effects on

the Koine Greek.


Suggested Classification Systems

          How, then, did Classical Greek scholars classify conditional

sentences? In general, they seemed to follow one of three systems.

C. D. Chambers outlines them as follows:


          1 A. T. Robertson, A Short Grammar, p. 6.

          2 Dana and Mantey, Grammar, p. 6.

          3 Ibid., p. 6.



          There are three possible ways of classifying conditional

      sentences, viz. (i) by time, (ii) by fulfillment, (iii) by form.

     The first is the system of Prof. Goodwin, the second is proposed by

     Mr. Donovan . . . and the third that of Mr. Sonnenschein.l

Though the situation is not as absolute as Chambers suggests, his

comments serve as a useful guide to the discussion of conditional

sentences in Classical Greek.


Classification According to Time

          The classification most familiar to American students of

Classical Greek is that of William Watson Goodwin. As Professor of

Greek Literature at Harvard University (1860-1901) he exerted signi-

ficant influence on Greek studies in the United States. His first major

book, Syntax of the Moods and Tenses of the Greek Verb, was published in

1860 when he was twenty-nine years of age. Ten years later he published

An Elementary Greek. Grammar.2  The importance of these works may be

judged by the fact that both are still in print. His influence

extended into New Testament studies through the work of Ernest DeWitt


          The statement of the system.--Goodwin sets forth his system in

terms of past, present and future conditions:


          1 C. D. Chambers, "The Classification of Conditional Sentences,"

The Classical Review, 9:2 (May, 1895), pp. 293-294.

          2 Chalmers G. Davidson, "Goodwin, William Watson," Dictionary

American Biography, Vol. IV, edited by Allen Johnson (New York: Charles

Scribner's Sons, 1957), pp. 411-413.




          The most obvious natural distinction is that of (a) present

      and past conditions and (b) future conditions. Present and past

      conditions (a) are divided into two classes by distinguishing (1)

      those which imply nothing as to the fulfillment of the condition

     from (2) those which imply that the condition is not or was not

     fulfilled. Future conditions (b) have two classes (1, 2), distin-

     guished by the manner in which the supposition is stated. Class 1

     of present and past conditions is further distinguished on the

     ground of the particular or general character of the supposi-

      tion . . . .1

          Goodwin also includes the concepts of (a) fulfillment or non-

fulfillment and (b) particular and general characteristics as other

features by which conditional sentences may be classified. The first

leads him to identify conditional clauses as (1) those which imply

nothing as to the fulfillment of the condition, (2) those which imply

fulfillment of the condition and (3) those which imply the nonfulfillment

of the condition.

          He explains the particular and general characteristics:

          A particular supposition refers to a definite act or to several

      definite acts, supposed to occur at some definite time (or

      times) . . . .

          A general supposition refers indefinitely to any act or acts

      of a given class which may be supposed to occur or to have

      occurred at any time . . . .2

          The form of the conditional sentence may serve as a guide to

its identification and classification under these headings. A later

edition of his grammar summarizes this:

          I. Present and past suppositions implying nothing as to ful-

               fillment or condition:

                    (a) Chiefly particular:

                              (protasis) ei] with indicative

                              (apodosis) any form of the verb


          1 Goodwin, Syntax of Moods, p. 139.

          2 Ibid., p. 141.


          (b)      General:

                    1.  (protasis) e]a<n with subjunctive

                         (apodosis) present indicative

                    2.  (protasis) ei] with the optative

                         (apodosis) imperfect indicative

II.  Present and Past suppositions implying that the condition is

      not fulfilled:

          (protasis) ei] with past tense of indicative

          (apodosis) past tense of indicative with a@n

III. Future suppositions in more vivid form:

          (protasis) ei] with subjunctive (sometimes ei] with future


          (apodosis) any future form

IV.  Future suppositions in less vivid form:

          (protasis) ei] with optative

          (apodosis) optative with a@n1

Smyth also adopts this system of classification.2

          The evaluation of the system.--The first point in evaluating

Goodwin's system is that of time. He seems to make time a basis of

classification when absolute time is of secondary importance in the

Greek verb system. Goodwin himself notes that relative time is far more

prominent in Greek verbs than in English: "It is a special distinction

between the Greek and the English idioms, that the Greek uses its verbal


          1 William Watson Goodwin, Greek Grammar, revised by Charles

Burton Gulick (Boston: Ginn and Company, 1930), pp. 295-296.

          2 Smyth, Grammar, pp. 513-516.



forms much more freely to denote merely relative time."1 A. T. Robertson,

noting that absolute time is limited to the indicative mood, warns that

"even in the indicative the time element is subordinate to the kind of

action expressed."2  While his reference to "kind" of action may be

questioned, his observation on time is helpful.

          But is Goodwin basing his classification on time as indicated

by the verb itself, or on absolute time as indicated by the sentence as

a whole? The examples he offers as illustrations seem to place the

emphasis upon the verb rather than the syntax. Thus he presents: "Ei]  

pra<ssei tou?to, kalw?j e@xei, if he is doing this, it is well;  ei] pra<ssei

tou?to, h[marthke<n, if he is doing this, he has erred; ei] pra<ssei tou?to,

kalw?j e@cei, if he is doing this, it will be well."3  One may well argue

that the syntax of a verb does assign it absolute time in any given

context. But Goodwin does not stress the role of syntax in establishing

the time of his verb. One must therefore question any system of

classification which makes absolute time as found in the verb itself a

foundation criterion.

          Second, his distinction between particular and general conditions

may be questioned as an objective standard of classification. Goodwin

argues that


          1 Goodwin, Syntax of Moods, p. 8.

          2 A. T. Robertson, Grammar, p. 825.

          3 Goodwin, Syntax of Moods, p. 139.



          When the apodosis has a verb of present time expressing a

      customary or repeated action, the protasis may refer (in a general

      way) to any act or acts of a given class which may be supposed to

      occur at any time within the period represented in English as

      present. Thus we may say:--

          Ea@n ti<j kle<pth, kola<zetai, if (ever) any one steals, he is

(in all such cases) punished . . . .1

          Goodwin's concept of a present general condition seems to fit

conditional sentences in the New Testament, such as I John 1:7:  e]a<n de>

e]n t&? fwti> peripatw?men. . . , koinwni<an e@xomen - but if (whenever) we

walk in the light . . . we have (in such cases) fellowship. But what

about conditions such as Luke 5:12? Here the apodosis uses a present

tense, but the condition must be considered a particular one, limited

to the historical situation:  ku<rie, e]a<n qe<lhj, du<nasai< me kaqari<sai -

Lord, if you are willing you are able to cleanse me.

          What is the difference between these two sentences? Simply

that the apodosis of I John 1:7 contains a present tense verb which

expresses a "customary or repeated action," while that of Luke 5:12

does not. There can be no question that the condition in I John 1:7

states a general situation that is presently true for all believers,

but such identification depends upon the interpretation of the action

represented by the verb. Could it be possible for interpreters to

disagree over the interpretation of a given verb? Yes, it could.

Should the basis of classification be a point that is interpretative

in nature? It seems reasonable to answer in the negative. Since, then,

there is no objective way of determining if a verb is referring to a


          1 Goodwin, Syntax Mood's, p. 141.



general or a particular act, the final determination becomes one of

interpretation rather than form. The concept of general versus particular

may serve as an interpretative guideline, but it is not distinguished by

form. "That point [of present or general conditions]," writes A. T.

Robertson ,"has no bearing on the quality of the condition."1  Though

several modern New Testament grammarians continue this terminology,

it must be questioned as a criterion to the objective analysis of

conditional sentences.

          Another point to consider in evaluating Goodwin's classifica-

tion system is his concept of fulfillment or non-fulfillment as found

in the condition. Robertson has particular problems with Goodwin's

concept that conditions employing the indicative mood in the protasis

imply nothing as to the fulfillment of the condition. This, as

Robertson sees it, violates the very nature of the indicative mood:

      The words to which I object, besides "particular," are "implying

      nothing as to the fulfillment of the condition." This condition

      pointedly implies the fulfillment of the condition. It is the

      condition of actuality, reality, Wirklichkeit, and not mere

      "possibility" as Farrar has it . . . a la Goodwin.2

Robertson claims that Goodwin "confuses the 'fact' with the 'statement'

of the fact."3 This seems a bit harsh, for Goodwin himself writes:

"The Greek has no form implying that a condition is or was fulfilled,

and it is hardly conceivable that any language should find such a form


          1 Robertson, Grammar, p. 1006.

          2 Ibid., p. 1006.

          3 Ibid.,p. 1006.



necessary or useful."1  This would amount to a direct statement, for

if the condition is fulfilled, then the results are realized. Such a

statement would not be a conditional statement at all, but a statement

of actuality. While the specific topic of reality in conditional

sentences will be examined in detail later, the important point is that

Robertson had serious doubts about the validity of Goodwin's classifica-

tion scheme.

          Another minor point of criticism leveled against Goodwin is

that of terminology. J. W. Roberts notes that "Others have attacked

Goodwin's terms 'more' and 'less vivid' as describing the significance

of his third and fourth class conditional sentences," but gives no

supporting references.2  Robertson did not use this terminology, and

those who followed have also set it aside. Some contemporary grammarians,

though, are returning to it. Both Robert W. Funk and William S.

La Sor speak of "vivid" and "less vivid" concepts when discussing

conditional sentences in their grammars. This point is not foundational

to the analysis of Goodwin's system, and will be discussed later.


          1 Goodwin, Syntax of Moods, p. 140.

          2 J. W. Roberts, "The Use of Conditional Sentences in the Greek

New Testament as Compared with Homeric, Classical and Hellenistic

Uses," unpublished Ph.D. dissertation, the University of Texas,

1955, p. 20.

          3 Robert W. Funk, A Beginning-Intermediate Grammar of Hellenistic

Greek (Missoula, Montana: The Society of Biblical Literature, 1973),

p. 684; and William Sanford La Sor, Handbook of New Testament Greek,

vol. II (Grand. Rapids: Wm. B. Eerdmans Publishing Company, 1973),

pp. 221-225.



          In summary, then, Goodwin is the main proponent of Chamber's

first suggested criterion for classifying conditional sentences: Time.

The main point of Goodwin's system is the classification of these

sentences into past, present and future conditions. Some of these

categories are further divided into "particular" and "general" condi-

tions, and some of these are subdivided by "vividness."

          Criticism of his system has focused on (1) his use of time as

a main dividing point, (2) the characteristics of particular and

general, (3) the implication of fulfillment and (4) his terminology.

The majority of Koine grammarians today, operating under the influence

of A. T. Robertson, do not follow Goodwin's system, though some show

signs of returning to it.


Classification According to Fulfillment

          The second criterion suggested by Chambers for classifying

conditional sentences is that of fulfillment, i.e. cataloging them

according to the fulfillment or non-fulfillment of the condition.

          The statement of the system.--Herbert Wier Smyth follows this

system, noting that conditional sentences may be classified according

to form and function. Among the functions he lists is "fulfillment or

non-fulfillment."1  J. Donovan also champions this method of classifica-

tion, and his argument is worth pursuing. He uses a book review essay

as an opportunity to argue his case in The Classical Review. The grammar


          1 Smyth, Grammar, p. 514.



he is reviewing argues for the position of classification by form, and

Donovan rejects this approach. He argues that different meanings may

have the same form. He concludes by writing: "What is wanted is not that

there should be a search for 'would be' or 'should be' or 'might be'

or other variations; but recourse should at once be had to the universal

canon of fulfillment or non-fulfillment."1

          Further, recognizing that his position had already been

challenged, he adds a note of defense:

      After the campaign recently conducted in the pages of this Review

      against the very principle of fulfillment as a basis of classifica-

      tion, one point is now clear, if it was not so already, namely,

      that to be regarded as unfulfilled, a condition need not necessarily

      be so actually, but that it is enough that it should be assumed

      to be such.2

Thus Donovan elevates the concept of fulfillment or non-fulfillment to

the status of a "universal canon" for the classification of conditional


          The evaluation of the system.--Donovan's review brought a rapid

response from Chambers who defended the position in question. His

rebuttal begins with a summary of Donovan's principle:

      Therefore the universal canon resolves itself into this: Conditions

      are to be divided into (i) those which imply or assume without

      implying that the condition is not fulfilled, and (ii) those which

      do not assume or imply that the condition is not fulfilled.3


          1 J. Donovan, "Sonnenschein's Greek Grammar," The Classical  

Review, 9:1 (January, 1895), p. 64.

          2 Ibid., p. 64.

          3 Chambers, "Classification," pp. 293-294.



He then offers three serious objections to Donovan's position. First,

Chambers observes that this produces an imbalance in grammar. Condi-

tions implying non-fulfillment are relatively rare in the language and

should not be the basis of classification. Second, the terminology is

awkward at best and hardly fits the need of the beginning composition.

Since Chambers is speaking of English to Greek composition, his comment

has little reference to this study. Third, and more to the point,

Donovan's scheme does not fit all cases. The majority of grammarians

have not followed his suggestions. The concept of fulfillment does, as

Goodwin notes, play a role in our understanding of conditional sentences,

but it does not provide a sufficiently applicable standard upon which to

classify them.1


Classification by Form

          The third criterion Chambers suggests is classification by the

form of the conditional sentence. Among the classical grammarians who

have followed this approach is E. A. Sonnenschein, professor of Greek

and Latin at Birmingham University.2


          The statement of the system.--Sonnenschein writes:

          To me the ordinary forms of Conditional Sentences, whether in

      Latin, Greek or a modern language, present themselves in two great


          1 Goodwin, Syntax of Moods, p. 139.

          2 S. B. Sedwick, "Sonnenschein, Edward Adolf," Dictionary of  

National Biography, 1922-1930, edited by J. R. H. Weaver (London:

Oxford University Press, 1961), p. 796.



     classes, the distinction between which is marked by certain well-

     defined differences both of meaning and of linguistic form.1

Form, for him, helps determine the meaning, for in his classification

"there is a coincidence between distinctions of form and distinctions of

meaning."2 Chambers supports this principle, noting that "It is

axiomatic that the division by form, and not by sense, is the truly

scientific one, because like forms must (originally at least) have like

meanings, but like meanings need not have like forms."3

          Smyth notes that several possibilities exist with regard to

classification according to form:

          Classified according to form, all conditional sentences may be

          arranged with regard to the form of the protasis or of the apodosis.

          Protasis:  ei] with the indicative.

                    e]a<n (rarely ei]) with the subjunctive.

                    ei] with the optative.

          Apodosis: with a@n, denoting what would (should) be or have been.

                    without a@n, not denoting what would (should) be or

                    have been.4

          Which of these possibilities should be followed if conditional

sentences are to be classified by form? "Ought we to classify according

to the Protasis (Subordinate Clause) or according to the Apodosis


          1 E. A. Sonnenschein, "Horton-Smith's Conditional Sentences,"

The Classical. Review, 9:2 (April, 1895), p. 221.

          2 Ibid., p. 221, italics added .

          3 Chambers, "Classification," p. 294.

          4 Smyth, Grammar, p. 513.



(Principal Clause), or according to both at once?"1  Sonnenschein opts

for the Apodosis as the standard. Roberts outlines his system so:

          I.  Type One: Sentences without a@n in the apodosis: the protasis

                    expressing no implication of fulfillment.

          ei] with a Past Indicative expressing what was

          ei] with a Present Indicative expressing what is

          e]a<n with a Subjunctive expressing what is or what will be

          Also e]a<n h] e@stai "If A is B," followed by a command or wish.

II.  Type Two: Sentences with ay in the apodosis: the protasis

          expressing some sort of implication as to fact or fulfill-


          ei] with the Optative--Optative with a@n expressing what would be

          ei] with a Past Indicative--Past Indicative with a@n expressing

                    what would be

III.      Type Three:  ei] ei@h . . . e@stai (or e@sti) expressing in the

                    protasis some sort of mental reservation.2

          Evaluation of the system.--Donovan, as indicated above, claims

that this system leads to more problems than solutions in application.

Even Chambers wonders "whether this was the original principle of

division, or only its accidental final result."3 The basic problem is

the occasional divergence between form and meaning. Although the use of

form does offer an objective standard of classification, is it reasonable

to pursue it when it does not coincide with the real world? As will be shown

Sonnenschein is on the right track, but the emphasis requires a slight


          1 Sonnenschein, "Horton-Smith," p. 220.

          2 Roberts, "Conditional Sentences," p. 12.

          3 Charmers, "Classification," p. 294.



shift in order to be workable.


Classification by Determination

          Chambers omitted a fourth basis of classification of conditional

sentences, that of determination. This approach, popularized by B. L.

Gildersleeve and followed by many other grammarians, was first published

in 1876 and again in 1882.1  It seems strange that Chambers, writing in

1895, was unaware of it.

          The statement of the system.--Gildersleeve states his case so:

          In common with most grammarians, I divide the conditional

     sentences into four classes, for which I have been in the habit of

     using the designations "Logical," "Anticipatory," "Ideal,"

     "Unreal." If nothing more can be said in behalf of this nomenclature

     than that it saves time, something at least has been said; and I am

     glad to learn that a part of this nomenclature, as applied to the

     Latin language, has found favor among teachers. Logical, Ideal, and

     Unreal conditions occur in Latin also. The Anticipatory is

     peculiar to Greek.2

          He describes the Logical Condition as one that

     . . . states the elements in question. It is used of that which

    can be brought to the standard of fact; but that standard may be

    for or against the truth of the postulate. All that the logical

    condition asserts is the inexorable connexion [sic] of the two

    members of the sentence.3


          1 B. L. Gildersleeve, "On ei] with the Future Indicative and ean  

with the Subjunctive in the Tragic Poets," Transactions of the American

Philological Association, 7:1 (January, 1876), pp. 2-23; and “Pindaric

Syntax,” pp. 434-445.

          2 Gildersleeve, "On ei]," pp. 5-6.

.         3 Gildersleeve, "Pindar," p. 435.


This type of condition corresponds to Robertson's First Class condition.1

          The Anticipatory Condition involves e]a<n in the protasis, and thus

corresponds to Robertson's Third Class condition. Interestingly enough,

Gildersleeve agrees with Goodwin in his concept of particular and

general ("generic"), noting that "The anticipatory condition is

particular or generic according to the character of the apodosis . . .,

just as any other conditional sentence."2  The key element is the

use of the present indicative in the apodosis. Such a condition "is

regularly generic."3

          Gildersleeve's Ideal Condition employs the optative mood and

"seems to have been developed out of the wish, just as the anticipatory

was developed out of demand."4 This corresponds to Robertson's Fourth

Class condition. Since the New Testament has no complete sentence of

this type, his comments on it are beside the point of this study.

          His fourth type of condition is termed the Unreal Condition,

corresponding to Robertson's Second Class condition. Rather than use

the term non-fulfillment as does Goodwin, he speaks of it as


      The Unreal. Condition, 'the hypothesis contrary to fact,' seems

      to be related to the hopeless wish, as the ideal condition to the

      wish pure and simple . . . . A wish may be madly impossible, but


          1 Robertson, Grammar, p. 1007.

          2 Gildersleeve, "Pindar," p. 436.

          3 Ibid., p. 435.

          4 Ibid., p. 436.



     if it belongs to the domain of the future it is optative. Now the

     hopeless wish is hopeless because it is futureless . . . .1

In simple language, this condition states a condition as though there

is no hope whatsoever of its being fulfilled. It implies its non-


          Thus Gildersleeve divides conditional sentences into two broad

catagories: the first containing those which imply something about the

determination of the condition, and the second containing those which

imply nothing about its determination. The first catagory is subdivided

into two classes: those which imply positive fulfillment of the condi-

tion and those which imply negative or non-fulfillment of the condition.

The second catagory also is subdivided into two classes: those

conditions with a greater degree of probability and those with a

lesser degree of probability.

Roberts summarizes this in outline form:

I.        Condition determined

          A.      As fulfilled - the Simple or Logical Condition


                    Protasis: ei] with an indicative present


                    Apodosis: Any form of verb

          B.       As unfulfilled - the Unreal Condition

                    Protasis: ei] with a past indicative

                    Apodosis: Past indicative with a@n


          1 Gildersleeve, "Pindar," p. 437.



II.  Condition undetermined

          A. With greater prospect of fulfillment - The Anticipatory


                    Protasis: e]a<n with the subjunctive

                    Apodosis: Usually future, except for the general, which has

                              the present indicative

          B. With less prospect of fulfillment - The Ideal Condition

                    Protasis: ei] with the optative

                    Apodosis: Optative with a@n1

          The comparison of the system.--A comparison of this outline with

that of Goodwin's system on pages 12-13 or with the chart in Smyth's

grammar will indicate Gildersleeve's points of departure.2  These may

be listed as:

          1. No attempt is made to utilize the time of the condition as a

guide to the classification of the sentence.

          2. No attempt has been made to divide them into particular

or general on the basis of the protasis. Each of these may, in

Gildersleeve's opinion, be particular or general, depending on the

apodosis, but that is not a basis of classification.3

          3. Mood, rather than tense is emphasized as one of the important

features of the protasis.


          1 Roberts, "Conditional Sentences," p. 22.

          2 Smyth, Gammar, p. 516.

          3 He does approve of such a distinction as a guide to interpre-

tation. See "On ei]," p. 7.



          4. No attempt is made to distinguish general and particular

conditions by form, though he does recognize that the Anticipatory

condition (e]a<n with the subjunctive in the protasis) is more often than

not a general condition.

          5. The Future Condition (called by Smyth the "Future Emotional"1)

is identified as a simple or logical condition on the basis of the

indicative mood.

          This system has been followed by a significant number of Koine

grammarians, including men such as Winer,2 Buttman,3 Robertson,4 and


          The evaluation of the system.--There are three points which need

to be considered in evaluating his system. First, the above-mentioned


          1 Smyth, Grammar, p. 516.

          2 G[eorge] Benedict] Winer, A Grammar of the Idiom of the New

Testament, seventh edition, revised by Gottlieb Lunemann, translated

and edited by J. H. Thayer (Andover, Massachusetts: Warren F. Draper,

Publisher, 1893), p. 291.

          3 Alexander Buttmann, A Grammar of the New Testament Greek,

translated and edited by J. H. Thayer (Andover: Warren F. Draper,

Publisher, 1873), p. 220.

          4 Robertson, Grammar, p. 1004.

          5 F[riedrich] Blass and Albert] Debrunner, A Greek Grammar of

the New Testament and Other Early Christian Literature, translated and

revised by Robert W. Funk (Chicago: The University of Chicago Press,

1961), pp. 188-189.



point regarding present and general conditions needs to be summarized.

Gildersleeve maintains that any conditional sentence may be particular

or general "according to the character of the apodosis."1  The

particular character he looks for is the use of the present tense. This

is true, according to him, of all conditions, including the Logical and

Anticipatory conditions:

     Hence when [the logical condition] has its apodosis in the present,

     it has a double meaning, which adapts it admirably to personal

     argument. So especially when the form ei] tij is used, which may

     point either to a definite or to an indefinite person, the Logical

     condition is a two-edged sword, often wielded in the keen encounter

     of Attic wit. But as the e]a<n conditional with a present indicative

     apodosis is regularly generic, it is not without reason that this

     form should be preferred, when distinctly generic action is to be


          New Testament examples which illustrate his point include

Matthew 19:10 (ei] ou!twj e]sti<n h[ ai]ti<a tou? a]nqrw<pou meta> th?j gunaiko<j, ou] sumfe<rei gamh?sai - if this is the case of the man with his wife,

it is better not to marry), a logical condition which is obviously

a general one, and John 11:9 (e]a<n tij peripath? e]n t^? h[me<r%, ou]

prosko<ptei - if anyone walks in the day he does not stumble), an

anticipatory condition which is also general. Whether this holds as

a uniform rule (present tense in apodosis = a general condition)

deserve detailed treatment and will receive it at a later point in

this study. For now, it is sufficient to note that Gildersleeve's

suggestion does fit some passages in the Koine of the New Testament.


          1 Gildersleeve, "Pindar," p. 435.

          2 Gildersleeve, "On ei," p. 6.



          The second point of evaluation also relates to a distinction

between Gildersleeve and Goodwin. Gildersleeve's system relegates the

present general condition of Goodwin to the anticipatory class because

of its subjunctive mood. Conditions employing future indicative verbs

are classed as simple or logical conditions. But what is the difference

between these two? Does not the future indicative convey the same concept

as the present subjunctive, for all practical purposes?

          Gildersleeve answers that there is a distinction between the

two, one which he feels has been overlooked by many grammarians. First,

he notes, the normal pattern for future conditions is e]a<n with the

subjunctive: "The fact then is patent enough to every one who will be

at pains to count, that for model Greek prose e]a<n with the subjunctive

is preferred to ei] with the future indicative."1 The reason for this,

he writes,

      . . . seems to be to a considerable extent the greater temporal

     exactness, the same greater temporal exactness which has wholly

     displaced the future indicative with the temporal particles, the same

     greater temporal exactness which has given so wide a sweep to the

     optative with a@n as a sharper form of the future.2

          If e]a<n with the subjunctive is the normal form in Attic prose,

what is the role of ei] with the future indicative?  It shows

     . . . a certain coldness, a certain indifference; and this added to

     the general rigor of the logical condition, which faces fact in all

     its grimness, gives a stern, minatory, prophetic tone to the future


          1 Gildersleeve, "On ei]," p. 9.

          2 Ibid., p. 9.



     indicative, which commentators and grammarians have noticed, but

     noticed only in passing.1

          Though Gildersleeve is speaking of conditions in Attic Greek,

his observations give an added dimension to the force of such conditions

in the New Testament, such as Matthew 6:23 (e]a<n de> o[ o]fqalmo<j sou

ponhro>j h#, o!lon to> sw?ma< sou skoteino>n e@stai - but if your eye is evil,

your whole body will be darkness) and Luke 13:3 (a]lla< e]a<n mh> metanoh?te,

pa<ntej o[moi<wj a]polei?sqe - but unless you repent, you will all likewise

be destroyed).

          Thus Gildersleeve uses the mood of the verb as the guiding

principle of his classification. The present general conditions as

identified by Goodwin simply follow the pattern of Attic prose and should

be considered as a type of anticipatory or future conditions. Further,

the use of the future indicative not only classes the condition as a

logical condition, but stresses the inescapable nature of the apodosis.

          A third point arises over the distinction between the indicative

and subjunctive moods. Contrary to Gildersleeve, Goodwin maintains that

there is no distinction between these two moods in conditional sentences

except that of time. He devotes an entire paper to the defense of his

position and offers the following observation:

          The idea of "possibility" or something of the kind being attached

      to the subjunctive, it was naturally supposed that the simple

     indicative in protasis must have a corresponding idea at its

     foundation, and that of "certainty" or "reality" has generally been

     assigned to it.2


          1 Gildersleeve, "On ei]," p. 9.

          2 "William Watson Goodwin, "On the Classification of Conditional

Sentences in Greek Syntax," Transactions of the American Philological  

Association, 6:1 (March, 1873), pp. 61-62.



          He then gives several reasons for questioning that this

distinction holds up in conditional sentences. In concluding his

argument he writes:

     After the most careful study that I have been able to give to the

     subject, and especially after a comparison of several thousand

     classic examples, I am convinced that no such principle [of

     distinction] can be found. Every example that I have met with has

     only confirmed the opinion, which I can now express with the

     greatest confidence, that there is no inherent distinction between

     the present indicative and the present subjunctive in protasis

     (between ei] boule<tai and e]a<n boulh?tai) except that of time.1

          Robertson, writing some years later, defends Gildersleeve's

position against Goodwin by pointedly hinging the meaning of the logical

condition (ei] with the indicative in the protasis) on the significance

of the indicative mood.2  This mood, according to him, is characterized


     . . . the "modus rectus." It does express "l'affirmation pure

     et simple." The indicative does state a thing as true, but does not

     guarantee the reality of the thing. In the nature of the case only

     the statement is under discussion. A clear grip on this point will

     help one all along. The indicative has nothing to do with reality

     ("on sich"). The speaker presents something as true. Actuality is

     implied, to be sure, but nothing more. Whether it is true or no is

     another matter.2

          Concerning the subjunctive mood, Robertson notes two things.

First, it is probably impossible to identify a single root-idea for

this mood. He accepts Brugmann's identification of "three uses of the

subjunctive (the volitive, the deliberative, the futuristic."3  Thus


          1 Goodwin, "Classification," pp. 64-65.

          2 Robertson, Grammar, p. 915.

          3 Ibid., pp. 926-927.



the subjunctive does not necessarily imply a specific time.

          Second, there is a close connection between the aorist sub-

juctive and the future indicative:

      These [the aorist subjunctive and the future indicative] are closely

      allied in form and sense. It is quite probable that the future

      indicative is just a variation of the aorist subjunctive . . . .

      The subjunctive is always future, in subordinate clauses relatively

      future. Hence the two forms continued side by side in the language.

      There is a possible distinction. "The subjunctive differs from the

      future indicative in stating what is thought likely to occur, not

      positively what will occur." [quoting Thompson, A Syntax of Attic

      Greek, p. 133].1

          Thus Robertson offers support for Gildersleeve's position by

holding, first, to a uniform distinction between indicative and sub-

junctive moods, and, second, to a non-chronological significance for

the subjunctive mood itself. The particular case of the future

indicative and the aorist subjunctive may show a blurring of this

otherwise sharp distinction, but Robertson will not concede confusion

in the essential modal significance, even in conditional sentences.

          It would seem, then, that the major objections against

Gildersleeve's system raised by Goodwin can be answered. Indeed, some

of the answers are directly related to the objections raised against

Goodwin's own approach. Most Koine grammarians have been convinced of

the superiority of Gildersleeve's approach and have adopted it, via

Robertson, with some notable exceptions.


          1 Robertson, Grammar, pp. 924-925.



Summary of Classical Greek Classification Systems

          This somewhat detailed review of the work of Classical gram-

marians has shown that there is, in truth, significant disagreement

among them. Goodwin classifies according to time, Donovan according to

fulfillment or, non-fulfillment, Sonnenschein and Chambers according to

form, and Gildersleeve according to determination. In terms of a

majority vote, classical grammarians usually follow Goodwin and Koine

grammarians generally follow Gildersleeve (or Robertson who follows



                        Conditional Sentences in Koine Greek

          It is impossible to separate the advances in Koine grammar

from those of Classical Greek, for they have moved hand-in-hand. At

times, Koine grammarians took the work of the Classical scholars and

brought it directly into New Testament studies, as Burton did with

Goodwin's classification of conditional sentences. Robertson is

representative of those Koine scholars who were conversant with Classical

studies, but saw fit to reject some of them and sharpen the focus of

others, as he did with Gildersleeve's work on these sentences.

          Nigel Turner has done the historian of grammatical studies a

great favor by publishing a chronological bibliography of all major

Greek works, beginning with the first New Testament grammar published

in 1655 and ending with the latest edition of Bauer's Worterbuch in



          1 Nigel Turner, A Grammar of New Testament Greek. Volume III:

Syntax (Edinburgh: T. & T. Clark, 1963), pp. vii-x.



          This discussion will limit itself to the major Koine grammarians

appearing on his list, including those who have published since he

compiled his catalog. For organizational purposes they will be divided

into "Early" and "Late" grammarians, with A. T. Robertson being the

dividing point. Just as the survey of Classical grammarians illustrated

the wide range of opinion on conditional sentences in the classic

writings, so this survey will demonstrate that Koine studies are in a

similar state of flux.


                              Early Grammarians

George Benedict Winer

          George Benedict Winer is identified by Dana and Mantey as the

first grammarian to operate with the concept that the Greek of the New

Testament is the Greek of the common people, not a special, Holy Ghost

language.1  In the preface to the Sixth Edition of his grammar Winer


          The fundamental error--the prw?ton yeu?doj--of the Biblical

     philology and exegesis to which we refer, consisted ultimately in

     this, that neither the Hebrew nor the language of the N.T. was

     regarded as a living idiom . . . designed to be used by men as the

     medium of intercourse.2

From this basis he develops his analysis of the Koine grammar, including

that of conditional sentences.


          1 Dana and Mantey, Grammar, p. 9.

          2 Winer, Grammar, p. v.



Statement of His Position

          Winer applies the results of the "enlightened philology" of his

day to New Testament Greek and offers this analysis of conditional or

hypothetical sentences:

          In HYPOTHETICAL sentences four kinds of construction occur:

          a. Pure condition: If thy friend come, give him my regards  

     (the case is put as real). Here the Indicative is used with

      ei]. . . .

          b. Condition with assumption of objective possibility (where

      experience will decide whether or not it is real): If thy friend

      should come  (I do not know whether he will come, but the result

      will show). Here e]a<n . . . with the Subjunctive is used.

          c. Condition with assumption of subjective possibility, the

      condition existing merely in thought:  If thy friend come (the case

      being conceivable and credible) I should be pleased to present my

      respects to him. Here ei] with the Optative is used.

          d. Condition believed to be contrary to the fact: were there a

     God, he would govern (but there is not). Had God existed from

      eternity, he would have prevented evil (but he has not existed).

      Here ei] with the Indicative is used,--the Imperf. in the first case,

      the Aor. or (much more rarely) the Plup. in the second . . .; in the

      conclusion likewise one of these two tenses.1


Summary of His Position

          Winer's optimistic statement that "the diction of the N.T. will

be found entirely in accordance with the preceding rules" indicates his

confidence in these four basic divisions.2 Though this has not been

fulfilled to the degree he predicted, Winer has anticipated the major

system used by Koine grammarians today. It is apparent that he is

following Gildersleeve's system, though using slightly different termin-

ology. He makes no reference to Gildersleeve, but it is possible that he


          1 Winer, Grammar, p. 291.

          2 Ibid., p. 292



was familiar with Gildersleeve's work. Like him, Winer distinguishes

between ei] and the future indicative verb in the protasis and e]]a<n with

the subjunctive, though he acknowledges that "such construction with the

Fut. would approximate most nearly to that with e]a<n . . . ."1  He

illustrates the difference between the two in this way:

      . . . but if all shall be offended in thee is a more decided

     statement than if all should be offended. In the latter, it is

     still altogether uncertain whether they will be offended; in the

     former, this is assumed as a future fact . . . .2

          Winer notes that "the exceptions to these rules in the N.T.

text are but very few, and occur for the most part only in particular

Codd."3  He discusses two types of exceptions: the use of ei] with the

subjunctive and e]a<n with the indicative. He does not discuss mixed

conditions, concessive particles or elliptical conditions.


Evaluation of His Position

          It is difficult at best to read many last-century grammarians

with any degree of comprehension. They wrote "for another eye, another

mind and another time." He further complicates the issue by trying to

illustrate Greek concepts from English. The difference between "If thy

friend come" and "If thy friend should come" hardly conveys the

difference between the indicative and subjunctive moods. His

explanatory comments are far more helpful than his examples, which tend

to confuse the points he tries to make.


          1 Winer, Grammar, p. 293.

          2  Ibid., p. 294.



          Winer did, however, establish a significant precedent by

considering the Koine Greek to be just that, the language of the common

man. By bringing the concept of simple observation and induction to

bear upon the New Testament he advanced the understanding of its

structure and pointed the way for others to follow.


Alexander Buttmann

          Alexander Buttmann followed the general rules of Winer in his

treatment of conditional sentences. Rather than expanding the rules at

length, Buttmann concentrated on the deviations found in the New Testa-

ment. By this time it was fairly evident that Winer's confident asser-

tion that all New Testament forms were covered by his four rules was

overstated. Buttmann's work, though, shows that Winer's four classifica-

tions are generally true and accurately describe the majority of

conditional sentences in the New Testament.


Statement of His Position

          Buttmann observed that the first two forms of conditional

sentences (ei] with the indicative and e]a<n with the subjunctive) are by

far the most frequent forms in the New Testament.1  He also stressed the

importance of mood as the determining factor in evaluating the kind of


      The difference between them [the two types of conditional

      sentences] . . . is plainly to be recognized in sentences where

      both are used in close proximity; as Gal. i. 8,9, where the


          1 Buttmann, Grammar, p. 220.



      hypothesis expressed in the 8th verse by e]a<n with the Subjunctive

      is resumed or repeated in the 9th verse with greater energy and

      definiteness by ei] with the Indicative. So in Acts v. 38, 39.1

Comparison of His Position

          Buttmann's analysis of the exceptions to Winer's rules centers

on two areas: the use of ei] and e][a<n with the indicative and subjunctive

moods respectively, and the use of a@n in conditions contrary to fact.

          Particles and moods.--His analysis of the particles and their

corresponding moods is one of the first treatments of the topic that

discusses the problem of variant readings.

     Of the first case, the use of ei] with the Subjunctive, we find, to

     be sure, accidentally . . . no example which is quite certain; for

     in some of them the readings vary, some are set aside by the MSS.

     (as Rev. xi. 5 [but cod. Sin. qelh?sh the second time]), some are

     capable of a special interpretation.2

He also lists I Corinthians 9:11 and Luke 9:13 as possible examples of

this exception.

          Regarding the use of  e]a<n with the indicative he notes that this

"is given so frequently, that it is to be eliminated as little from the

writings of the N.T. as of the Old."3  Buttmann does concede that most of

these examples may be questioned upon textual evidence, but argues

that the variants were introduced when the copyists altered the original

and more difficult indicative.


          1 Buttmann, Grammar, p. 220.

          2 Ibid., p. 221.

          3 Ibid., p. 222.



          It is, indeed, not to be denied that the instances in question

      almost disappear amid the multitude of those that are grammatically

      regular, and suspicion may also be raised by the circumstance that

      hardly a single passage with the Indicative is completely beyond

      question critically. Yet when we consider that in countless

      passages with the Subjunctive not the smallest variation is found

      (which would not be the case if the Indicative were chargeable solely

      to the copyists), it is far more probable that, where a diversity of

      readings occurs in such a number of instances, this fact results

     from the circumstance that the copyists, commentators, etc., early

     altered the Indicative which gave them offence.1

Specific examples will be considered in a later section of this study,

but Buttmann's consideration of variant readings marks a significant

advance in the detailed study of conditional sentences.

          Conditions contrary to fact.--The second area of deviation

Buttmann examined was the use of a@n in the fourth class of conditions:

conditions contrary to fact. He noted that the apodosis regularly

included a@n, but recognized that this was not an absolute principle.

He listed four rules to explain the disappearance of a@n from these


          a) When a@n has already been expressed previously in the same

      connection with another predicate. This instance, which often occurs

      in the classics and is found in the nature of the case, is acci-

      dentally not to be met with in the text of the N. T. . . .

          b) When the predicate (or the copula) to which it belongs is

      also dropped, as I Cor. xii. 19 . . . .

          c) Where the apodosis contains such a predicative term as e]dei>,

      kalo<n h#n, h]du<nato, etc. This omission . . . is so necessary

     according to Greek habits of thought, that it is only by way of

     concession to our usage that we can speak of supplying a@n.

          d) Lastly, a@n is dropped for rhetorical reasons: where,

      though the fact itself is impossible or improbable, the orator in

      the vivacity of his thought desires to represent it as actually


          1 Buttmann, Grammar, p. 222.



having occurred, or at least as almost taken place.1

          Most modern grammarians dispense with these rules, simply

noting that a@n usage is at best unpredictable. Robertson, for example,

simply states that "There is no principle involved in a@n, simply custom."2

Buttmann was willing to recognize that the particle may be absent from a

condition without upsetting the force of that sentence, and this is the

emphasis of modern grammars.


Evaluation of His Position

          Buttmann gets credit for a more thorough study of the conditional

sentence than that of Winer. He accepted Winer's four categories, but

noted several major exceptions to them, especially in the variant

readings. Buttmann and Winer both placed emphasis upon mood as the

determining factor in classifying conditional sentences rather than the

particles or tense. It remained for other grammarians to state this

principle in more formal terms.


Samuel G. Green

          Published around 1887, Samuel G. Green's Handbook of the Greek

Testament presents an analysis of conditional sentences that follows the

patterns of Winer and Buttmann.3


          1 Buttmann, Grammar, pp. 225-226.

          2 Robertson, Grammar. p. 1007.

          3 Samuel G. Green, Handbook to the Grammar of the Greek Testament

(London: The Religious Tract Society, 11886]), pp. 317-320.


Statement of His Position

          Briefly, Green identifies four kinds of conditions or hypotheses:

      a. The supposition of a fact.

       b.                    of a possibility.

       g.                      of uncertainty.

       d.                     of something unfulfilled.1

These four types of conditions are indicated by four distinct


          a. The conditional particle ei], if, with the Indicative in the

     protasis, assumes the hypothesis as a fact. The apodosis may have

     the Indicative or Imperative.

          b. Possibility or uncertainty with the prospect of decision, is

      expressed by e]a<n = ei] a@n (very rarely by ei] alone [He lists I Cor. 14:5,

      Phil. 3:12, and a few various readings, such as in Rev. 11:5, as

      examples.] with the Subjunctive in the conditional clause, and the

      Indicative or Imperative in the apodosis.

          g. The Optative in a conditional sentence expresses entire

      uncertainty--a supposed case. Here the particle ei] is always used.

          d. When the condition is spoken of as unfulfilled, the

       Indicative is used in both clauses, with the particle ei] in the

       protasis, and a@n in the apodosis.2

Evaluation of His Position

          As is evident, Green's system of analysis is the same as those

already discussed: four types of condition, each identified by a

particular combination of particles and moods, and each conveying a

different concept. Winer, Buttmann and Green all seek to analyze

conditional sentences in terms of form, especially that of mood. They

differ little from the popular scheme of Robertson.


          1 Green, Grammar, p. 317.

          2 Ibid., pp. 317-319.



Ernest DeWitt Burton

          Ernest DeWitt Burton was both a scholar and an administrator,

having served as the chairman of the Department of New Testament and

Early Christian Literature and later as the president of the University

of Chicago.1  In his major Greek work, Moods and Tenses of New Testament

Greek, he adopts Goodwin's analysis of conditional sentences and applies

it to the New Testament.2


Statement of His Position

          His specific position, following Goodwin, is:

          A. Simple Present or Past Particular Supposition. The protasis

     simply states a supposition which refers to a particular case in

     the present or past, implying nothing as to its fulfillment. The

     protasis is expressed by ei] with a present or past tense of the

     Indicative; any form of the finite verb may stand in the apodosis.


     John 15:20; ei] e]me> e]di<wcan, kai> u[ma?j diw<cousin, if they have persecuted

          me, they,will also persecute you.

     Gal. 5:18; ei] de> pneu<mati a@gesqe, ou]k e]ste> u[po> no<mon, but if ye are

          led by the Spirit, ye are not under the law. See also Matt. 4:3;

     Luke 16:11; Acts 5:39; Rom. 4:2; 8:10; Gal. 2:17; Rev. 20:15.

          B. Supposition Contrary to Fact. The protasis states a supposi-

      tion which refers to the present or past, implying that it is not or

      was not fulfilled.

          The protasis is expressed by ei] with a past tense of the

     Indicative; the apodosis by a past tense of the Indicative with an.

     John 11:21; Ku<rie, ei] h#j w$de ou]k a@n a]pe<qanen o[ a]delfo<j mou, Lord, if

          thou hadst been here, my brother would not have died.

     Gal. 1:10;  ei] e@ti a]nqrw<poij h@reskon, Xristou? dou?loj ou]k a@n h@mhn, if I

          were still pleasing men, I should not be a servant of Christ.

          See also John 14:28; Acts 18:14; Heb. 4:8; 11:15.


          1 Charles Thwing, "Burton, Ernest DeWitt," in Vol. II of

Dictionary of American Biography, ed. by Allen Johnson and Dumas Malone

(New York: Charles Scribner's Sons, 1958), pp. 341-342.

          2 Ernest DeWitt Burton, Syntax of Moods and Tenses in New Testament

Greek (Chicago: University of Chicago Press, 1897), p. 101.



          C. Future Supposition with More Probability. The protasis

      states a supposition which refers to the future, suggesting some

      probability of its fulfillment.

          The protasis is usually expressed by e]a<n (or a@n) with the

      Subjunctive; the apodosis by the Future Indicative or by some other

      form referring to future time.

      Matt. 9:21; e]a<n mo<non a!ywmai tou? i[mati<ou a]tou? swqh<somai, if I shall

          but touch his garments, I shall be made whole.

      John 12:26; e]a<n tij e]moi> diakonh? timh<sei au]to>n o[ path<r, if any man

          serve me, him will the Father honor.

     John 14:15; e[a<n a]gapa?te< me,  ta>j e]ntola>j ta>j e]ma>j thrh<sete, if ye

          love me, ye will keep my commandments. See also Matt. 5:20;

          I Cor. 4:19; Gal. 5:2; Jas. 2:15,16.


          D. Future Supposition with Less Probability. The protasis

      states a supposition which refers to the future, suggesting less

      probability of its fulfillment than is suggested by e]a<n with the

      Subjunctive. There is no perfect example of this form in the New


          The protasis is expressed by ei] with the Optative; the apodosis by

       the Optative with a@n.

      I Pet. 3:17; krei?tton ga>r a]gaqopoiou?ntaj, ei] qe<loi to> qe<lhma tou?

         qeou?, pa<sxein h@ kakopoiou?ntaj, for it is better, if the will of

          God should so will, that ye suffer for well doing than for evil

          doing. See also I Cor. 14:10; 15:37; I Pet. 3:14.

          E. Present General Supposition. The supposition refers to any

     occurrence of an act of a certain class in the (general) present,

     and the apodosis states what is wont to take place in any instance

     of an act of the class referred to in the protasis.

          The protasis is expressed by e]a<n with the Subjunctive, the

     apodosis by the Present Indicative.

     John 11:9; ea]<n tij peripath? e]n t^? h[me<r%, ou] prosko<ptei, if a man

          walk in the day, he stumbleth not.

     2 Tim. 2:5; e]a<n de> kai< a]qlh? tij, ou] stefanou?tai e]a<n mh> nomi<mwj

         a]qlh<sh, and if also a man contend in the games, he is not

          crowned, unless he contend lawfully. See also Mark 3:24; John

          7:51; 12:24; I Cor. 7:39, 40.

          F. Past General Supposition. The supposition refers to any

      past occurrence of an act of a certain class, and the apodosis

      states what was wont to take place in any instance of an act of the

      class referred to in the protasis.

          The protasis is expressed by ei] with the Optative, the apodosis

     by the Imperfect Indicative.



          There is apparently no instance of this form in the New Testa-


          In addition to these general classifications, Burton makes some

interesting observations regarding the specific classes. He notes that

in the first type

     . . . the Future Indicative may stand in the protasis of a conditional

     sentence of the first class when reference is had to a present

     necessity or intention, or when the writer desires to state not what

     will take place on the fulfillment of a future possibility, but merely

     to affirm a necessary logical consequence of a future event.2

This differs from Goodwin's position that the future indicative and the

subjunctive may have the same significance in conditional sentences.

          The third class of conditional sentences may also have these

constructions in the protasis: (a) ei] with the Subjunctive, (b) ei] or

e]a<n with the Future Indicative, and (c) ei] with the Present Indicative.3

Conditions of this last form are apparently first class conditions, but

"are distinguished by evident reference of the protasis to the future."4

          Concerning the fifth class, the Present General Supposition, he

notes that some conditions using ei] with the indicative "apparently

express a present general supposition," which does not fit his rule that

such conditions use e]a<n with the subjunctive.5  He explains that it is


          1 Burton, Syntax, pp. 102-106.

          2 Ibid., p. 103.

          3 Ibid.,  pp. 104-105.

          4 Ibid., p. 105.

          5 Ibid., p. 107.



difficult to distinguish between this form of a present general condition

and that of a simple condition:

      Yet in most New Testament passages of this kind, it is possible that

      a particular imagined instance in the present or future is before the

      mind as an illustration of the general class of cases . . . . It is

      scarcely possible to decide in each case whether the supposition was

      conceived of as general or particular

      Luke 14:26; ei@ tij e@rxetai pro<j me kai> ou] misei? . . . th>n yuxh>n

         e[autou?, ou] du<natai ei#nai< mou maqhth<j, if any man cometh unto

          me, and hateth not . . . his own life, he cannot be my disciple.

          Cf. John 1:51; 12:26; where in protases of apparently similar

          force e]a<n with the Subjunctive occurs, and the apodosis refers to

          the future.

      Rom. 8:25; ei] de> o! ou] ble<pomen e]lpi<zomen, di ] u[pomonh?j a]pekdexo<meqa,

          but if we hope for that which we see not, then do we with

          patience wait for it. See also Jas. 1:26.1

According to form, such conditions are simple conditions, but according

to interpretation, they may be considered a variety of present general

conditions. The distinction, it must be stressed, is one of interpreta-

tion, not form. Classification should be on an objective basis, such

as form, not upon a subjective one, such as interpretation.


Summary of His Position

          In addition to these details, Burton also considers many various

peculiarities of conditional sentences. He lists nine of these, including

(1) mixed forms, (2) multiple protases, each with its unique emphasis,

(3) the use of a participle, an imperative or other form of expression

"suggesting a supposition" to supply the protasis, and (4) the observation

that sometimes either the protasis or the apodosis may be omitted.2


          1 Burton, Syntax, pp. 107-108.

          2 Ibid., pp. 109-112.



          One final comment: Burton correctly identifies the assumption

of "reality" or "unreality" in these conditions as that of the speaker or

his hearers, not in the external situation:

          It should be observed that the titles of the several classes

      of conditional sentences describe the supposition not from the

      point of view of fact, but from that of the representation of the

      case to the speaker's own mind or to that of his hearers.1

Conditional sentences do speak of many things that are objectively true,

but the demonstration of their factuality lies in the external world,

not in the internal world of the conditional statement.


James Hope Moulton

          The Prolegomena to James Hope Moulton's A Grammar of the New

Testament was the first major grammar to utilize the newly discovered

evidence from the papyri.2  This work was followed by his Introduction to

the Study of New Testament Greek, a formal grammar.3  Moulton had

originally conceived of his major work, A Grammar of the New Testament,

in terms of three volumes. He published Volume I: Prolegomena in 1906,

and it quickly went through three editions in two years. He produced

the rough draft for parts I and II of the second volume, but was not

permitted to finish it. While at sea he died "in the Mediterranean,


          1 Burton, Syntax, p. 101.

          2 James Hope Moulton, A Grammar of New Testament Greek. Volume I:

Prolegomena, third edition (Edinburgh: T. S T. Clark, 1908), p. 4.

          3 James Hope Moulton, An Introduction to the Study of New Testa-

ment Greek (London: Charles H. Kelly, n.d.).



in April, 1917, a victim of the ruthless submarine campaign."1 Nigel Turner

continues the interesting story:

     His pupil, Dr. W. F. Howard, saw that volume through the press in

     parts, from 1919 to 1929, but before he had opportunity to lay many

     plans for Volume III he himself died in 1952; and then, on condition

     that he had the assistance of someone who would collect the necessary

     material, Dr. H. G. Meecham assumed responsibility for the syntax.

     It was on Dr. G. D. Kilpatrick's suggestion that I was permitted to

     help at this point, and we had done no more than compile a provisional

     bibliography when Dr. Meecham died in 1955. By the kind invitation of

     the publishers I then worked alone and broke the spell by living to

     complete Volume III.2

          Since he was anticipating two more volumes, Dr. Moulton did little

more than mention a few aspects of conditional sentences in the Prolegomena.

Fortunately, his Introduction to the Study of New Testament Greek gives

a relatively complete presentation of his views on the classification of

these clauses.


Statement of His Position

          First, the general comments from his Prolegomena will be

presented. To begin with, he notes that the distinction between ei] and e]a<n

     . . . has been considerably lessened in Hellenistic as compared with

     earlier Greek. We have seen that e]a<n can take the indicative; while

     (as rarely in classical Greek) ei] can be found with the subjunctive.3

          Regarding the constructional distinctions of conditional sentences,

Moulton makes the following observations:


          1 James Hope Moulton and W. F. Howard, A Grammar of New Testament

Greek. Volume II: Accidence and Word Formation (Edinburgh: T. & T. Clark,

1929), p. v.

          2 Nigel Turner, A Grammar of New Testament Greek. Volume III:

Syntax (Edinburgh: T. & T. Clark, 1963), p. v.

          3 Moulton, Prolegomena, p. 187.



     The differentation of construction remains at present stereotyped:

     ei] goes with indicative, is used exclusively when past tenses come

     in (e.g. Mk 326), and uses ou] as its negative; while e]a<n, retaining

     mh< exclusively, takes the subjunctive almost invariably, unless the

     practically synonymous future indicative is used.  Ea@n and ei] are both

     used, however, to express future conditions .         . . The immense

     majority of conditional sentences in the NT belong to these heads.1

          Moulton opts for Blass's principle as opposed to Goodwin's to

explain the use of the optative mood in these sentences:

          Meanwhile we may observe that Blass's dictum (p. 213) that

      ei] c. opt. form is used "if I wish to represent anything as generally

      possible, without regard to the general or actual situation at the

      moment," suits the NT exx. well; and it seems to fit the general

      facts better than Goodwin's doctrine of a "less vivid future"

      condition (Goodwin, Greek Grammar, 301).2

He specifically identifies Acts 8:31 as an example of a conditional

sentence employing a@n with the optative to which Goodwin's "less vivid"

form does not apply. 3

          However, he does follow Goodwin's general system for the overall

classification of conditional sentences. Three general classes are


      Simple Conditions in present or past time.

      Protasis, ei] with indicative; Apodosis, generally indicative, always

          without a]n.

These sentences merely join together a condition and a result without

any indication as to the probability or improbability of the condition.


          1 Moulton, Prolegomena, p. 187.

          2 Ibid., p. 196, note.

          3 Ibid., pp. 198-199.

          4 Ibid., p. 199.



      Unfulfilled Conditions in present and past time.

      Protasis, ei] with indicative, imperfect for present time, aorist for


      Apodosis, indicative with a@n, imperfect for present time, aorist for


      Future Conditions.

      Protasis, e]a<n with subjunctive (rarely indicative, or ei] with subjunc-


      Apodosis, future indicative, sometimes the imperative.1

          He classifies the optative condition, Robertson's Fourth Class

Condition, as a special form of the Future Condition, noting that its

full expression has vanished in the Koine and only parts of such conditions

appear in the New Testament.


                                           Modern Grammarians

Archibald Thomas Robertson

          Of all modern Koine grammarians, none has exerted the influence or

achieved the status of Archibald Thomas Robertson. As professor of New

Testament at the Southern Baptist Theological Seminary from 1890 to

1934 he helped form modern opinion about Koine Greek. With few exceptions

contemporary grammarians have adopted his terminology and viewpoint,

especially on conditional sentences. Thus his position requires detailed

study in order to fully comprehend the current majority view of

conditional sentences in the New Testament.


Statement of His Position

          His system of analysis basically follows that of Gildersleeve

in Classical Greek by identifying four types of conditions, each


          1 Moulton, Introduction to New Testament Greek, pp. 210-213.



determined by the mood of the protasis.

          His summary.--Robertson first summarizes the importance of mood

in conditions:

      The indicative mode in the condition always makes a clear-cut

      assertion one way or the other [fulfilled or unfulfilled]. If the

      subjunctive or the optative is used in the condition (protasis) a

      doubtful statement is made whatever may be the actual fact or truth

      in the case. By these modes of doubtful statement the condition

      puts it as doubtful or undetermined (not put in a clear--cut way).

      If the subjunctive is used, there is less doubt than if the optative

      is used, precisely the difference between these two modes of doubtful


          This distinction in mood (indicative = fulfilled or unfulfilled,

subjunctive = doubt, optative = more doubt) leads to the natural

conclusion that there are four types of conditional sentences:

          (a) First Class: Determined as Fulfilled (ei], sometimes e]a<n,

     with any tense of the indicative in condition. Any tense of the

      indicative in the conclusion).

          (b) Second Class: Determined as Unfulfilled (ei] and only past

     tenses of the indicative in condition. Only past tenses in the

     conclusion, usually with a@n to make clear the kind of condition


          (c) Third Class: Undetermined with Prospect of Determination

      (e]a<n or ei] with the subjunctive in the condition, usually future

      or present indicative or imperative in the conclusion, much variety

      in the form of the conclusion).

          (d) Fourth Class: Undetermined with Remote Prospect

     Determination (ei] with the optative in the condition, a@n and the

     optative in the conclusion).2


          1 A. T. Robertson and W. Hersey Davis, A New Short Grammar of

the Greek Testament (New York: Harper & Brothers, 1933), p. 349.

          2 Ibid., pp. 349-350.



          His Defence.--In defending this analysis, Robertson speaks

against the popular forms of classification, especially that of Goodwin.

First he rejects the concept of particular and general as a basic

principle of classification. Actually the concept of time was the key

principle of Goodwin, but the particular-general division was important.

In any event, Robertson points to the work of Gildersleeve and says of

Goodwin's distinction: "This is a false step in itself."1  He accepts

Gildersleeve's position that any condition may be particular or general,

depending upon the type of verb used in the protasis.

          Robertson then raises his next and most serious objection to

Goodwin and those following him: they refuse to recognize the basic

significance of the mood in conditions. Goodwin's first class of

conditions utilizes the indicative mood and, he says, "simply states a

present or past particular supposition, implying nothing as to the

fulfillment of the condition . . . .2  Robertson strongly objects to

this interpretation, claiming that "This condition pointedly implies the

fulfillment of the condition."3  Robertson hinges his argument on the

basic significance of the indicative mood which, he claims, has its usual

meaning in conditions as well as normal clauses. This is, as he says,

"the crux of the whole matter."4 Goodwin's classification seems to


          1 Robertson, Grammar, p. 1005.

          2 Goodwin, Moods and Tenses, p. 145.

          3 Robertson, Grammar, p. 1006.

          4 Ibid., p. 1006.


place emphasis on things other than mood, relegating it to a position

of lesser importance. For Robertson, mood is all-important. He

pursues his analysis on this assumption.

          One specific detail of his system needs further comment: the

first class condition labeled as "Determined as Fulfilled." Robertson

quotes Gildersleeve as identifying this condition as "the favorite

condition," though he is talking about classical poets, especially

Pindar.1  The question already raised by Robertson relates to the

degree of determination implied by the condition. Is the speaker

presenting the condition as something that is objectively true

(ei@ tij qe<lei o]pi<sw mou e@rxesqai, a]rnhsa<sqw e[auto<n - Since someone

does wish to come after me, let him deny himself. Luke 9:23)?  Or is the

speaker assuming the truth of the condition without committing himself

to a position one way or the other, such as might be done "for the

sake of the argument" (ei] de> a]na<stasij nekrw?n ou]k e@stin, ou]de> Xristo>j

e]gh>gertai - But if there is no resurrection of the dead, then Christ

is not raised. I Cor. 15:13)? This point will receive a more detailed

treatment later, but it is important to note that some commentators

have understood Robertson to say that the first class condition actually

affirms the objective reality of the condition. Perhaps his statements

could have been more precise, but common sense will suffice to show

that this cannot be the case in all situations. If it were, then

Christ would have been operating in the power of the Devil (Matt. 12:27)


          1 Robertson, Grammar, p. 1007.


and would not be resurrected (I Cor. 15:13). Yet writers continue to

to read objective reality into the first class condition. An extreme

example is the statement of Jerome Moore:

          The first class condition implies truth or reality. If . . .

      and it is true. Colossians 1:23 . . . is an example of this. The

      idea there is, "If ye continue in the faith grounded and settled,

      and ye shall!"  There is no doubt implied here. This is a condition

      of reality. No need to doubt the security of your salvation or any-

      one else's, for if here in the Greek does not imply doubt.1

          Were the situation not so serious, it would be funny.  No

one can guarantee the salvation of anyone else. Certainly Paul

challenged the Corinthians to test and examine themselves (2 Cor. 13:5).

Obviously some contexts permit the English "since" with its implication

of objective reality, but many, indeed, a majority, do not. To make

such claims is to ignore common sense, the teaching of Robertson and

the clear statements of Scripture.

          In his doctoral dissertation John Battle describes Robertson

as "difficult to read."2  Perhaps the wordiness of his Historical

Grammar led to some contradictory statements in the minds of some, but

in the first edition of his Shorter Grammar, he clearly states:

     This condition does assume the reality of the condition. Take

     Matt. 12:27. Christ did not cast out demons by Beelzebub, but

     in argument he assumes it. The indicative mode determines the

     condition as fulfilled, so far as the statement is concerned.3


          1 Jerome Moore, "Four Ways to Say 'If,'" The Baptist Bulletin

45:1 (June, 1979), p. 11.

          2 John A. Battle, Jr., "The Present Indicative in New Testament

Exegesis," unpublished Th.D. dissertation, Grace Theological Seminary,

1975, D. 170.

          3 A. T. Robertson, A Short Grammar of the Greek New Testament

(New York: A. C. Armstrong & Son, 1908), p. 151.


At times this assumption is parallel with the objective reality of

the statement, as in I Thessalonians 4:14 – ei] ga>r pisteu<omen o!ti

]Ihsouj a]pe<qanen kai> a]ne<sth . . . ; for if we believe that Jesus died and

rose again. In other contexts the assumption is counter to the

objective reality, as in Matthew 12:27 – kai> ei] e]gw> e]n Beezebou>l

e]kba<llw ta> daimo<nia . . . ; and if I by the power of Beelzebub cast out

the demons. In still other situations (probably the majority of those

in the New Testament) the assumption is neither parallel nor counter

to objective reality, for the reality cannot be determined from the

information at hand. This is the situation in Colossians 1:23 – ei@ ge

e]pime<nete t^? pi<stei . . .; if you continue in the faith. The best

English word to use in all three situations is "if," and all major

English translations of the New Testament uniformly translate the

first class condition this way.


Evaluation of His Position

          A. T. Robertson's analysis of conditional sentences has blazed

a trail that many have followed in New Testament studies. His

terminology has become almost universal, and his rejection of Goodwin's

system has been accepted by almost all who have followed him. Unfortun-

ately, some have taken Robertson too simply and have read verification

(or non-verification) of external or objective reality into his First

and Second Class conditions. One could wish that he had been more

precise in his statements, but such difficulties are no excuse for

the misuse of his concepts one finds in the literature.


William Douglas Chamberlain

          As a professor at Louisville Presbyterian Seminary, William

Douglas Chamberlain first published his Greek grammar in 1941: His

analysis is a concise summary of Robertson's work. Only a few remarks

need be made to indicate additional information he provides.


Statement of His Position

          Accepting Robertson's terminology, he affirms that in first class

conditions "The protasis has to do with the way the statement is made, and

not with the truth or falsity of it."1  Untrue conditions may be assumed

to be true for the sake of the argument. Matthew 12:27 is presented as

an example of this situation.

          In second class conditions he, like Robertson, identifies the

tenses used as past tenses: imperfect, aorist or pluperfect. It is

possible to have different tenses in the protasis and the apodosis, as

in John 14:28. While "the viewpoint is changed between the protasis and

the apodosis," the entire sentence is still a second class condition:

"These are not 'mixed conditions."'2

          He identifies the third class conditions as those which are

"stated as a matter of doubt, with some prospect of fulfillment."3

The fourth class is "even more doubtful than the third class."4  While


          1 William Douglas Chamberlain, An Exegetical Grammar of the Greek

New Testament (New York: The MacMillan Company, 1952), p. 195.

          2 Ibid., p. 197.

          3 Ibid., p. 198. 

          4 Ibid., p. 199. 


he finds no complete example of this condition in the New Testament, he

does identify fragments in I Peter 3:14, I Corinthians 15:37, Acts 17:27,

and Acts 27:39.1

          Chamberlain discusses two other aspects of conditional

sentences which are also mentioned in Robertson, though not in so

succinct a manner: Mixed Conditions and Elliptical Conditions. The

first involves a change in class of condition between the protasis and

the apodosis, for the "writer changes his viewpoint between the protasis

and the apodosis."2 Luke 17:6 is listed as an example.

          The second topic, elliptical conditions, involves conditional

sentences in which the apodosis is expressed and the protasis is simply

implied. He lists four ways this is accomplished:

          1. By the participle: Rom. 2:27.

          2. By a verb in the imperative mode: Mark 1:17.

          3. The protasis may be abbreviated to the vanishing point as with

      ei] mh> in the case of 'except': Mt. 11:27.

          4. The apodosis may be omitted: Luke 19:42.3

Chamberlain also notes that the Hebraistic use of ei] in oaths

(Mark 8:12), and its use to introduce direct questions (Acts 1:6) are

not conditional sentences.4


Evaluation of His Position

          Chamberlain's little grammar provides a very readable synthesis

of Robertson's position without going into the fine details of historical


          1 Chamberlain, Grammar, p. 199.

          2 Ibid., p. 199.

          3 Ibid., p. 199

          4 Ibid.,  p. 200.


analysis. Mood is the key factor in determining the type of conditional

sentence, and the indicative mood identifies sentences which present

the condition as true.


Charles Francis Digby Moule

          The English scholar C. F. D. Moule, publishing in 1953,

follows neither Robertson nor Goodwin in his analysis of conditional

sentences. He presents a unique system of both construction and

application, setting aside many of the chief dicta of other grammarians.

Statement of His Position

          He summarizes the various conditions under three headings:

          1. Past or present conditions, possible or actual.

          2. Recurrent or future conditions, whether real or hypothetical.

          3. Past or present conditions, only hypothetical.1

In outline form his system looks like this:

          1. Past or present conditions, possible or actual.

              Protasis:  ei] with the indicative in the appropriate tense.

              Apodosis: another indicative or its equivalent [an imperative,

              as in Col. 4:10, or conceivably a participle] in the appro-

              priate tense.

          2. Recurrent or future conditions, whether real or hypothetical.

              Protasis: ei] (or o!te) with a@n (making e]a<n, o[ta<n) with the

               subjunctive in the appropriate tense.

               Apodosis: Indicative or its equivalent [imperative or

               participle] in the appropriate tense.

          3. Past or present conditions, only hypothetical.


          1 C[harles] F[rancis] D[igby] Moule, An Idiom Book of New Testa-

ment Greek (Cambridge: Cambridge University Press, 1953), p. 148.


          Protasis: ei] with a past tense of the indicative.

          Apodosis: a past (but not necessarily the same) tense of the

          indicative, usually with a@n.1

          He sums up his concept of conditional sentences by stating:

          Thus the form of a conditional sentence is largely determined

     by two main factors—time (past, present, future) or Aktionsart  

     (instantaneous, protracted, recurrent, etc.) and the degree of

     reality (impossible, improbable, possible, probable, actual).2

Specific factors which help determine the mood of conditional sentences


     (a) Any past condition introduced by if must, in the nature of the

     case, be hypothetical, if not definitely unreal: otherwise there

     would be nothing conditional about the sentence. Therefore there

     appears to be no need to vary the mood, and it is regularly


     (b) But present, future, or recurrent conditions may vary widely

     in their degree of actuality: hence (perhaps) the variation in

     moods. In general, the Indic. represents certainty, while the

     Subj. represents something more hypothetical or uncertain.3

Moule thus recognizes the basic significance of these moods, but he

employs them in a unique system of analysis.


Evaluation of His Position

          Two observations may be made about this analysis. First,

Moule maintains that the apodosis is always in the indicative mood

regardless of the mood of the protasis. The specific examples will

be discussed in the next chapter, but this is an over-simplification, as

Galatians 5:25 demonstrates. Second and more important, this approach


          1 Moule, Idiom Book, pp. 148-149.

          2 Ibid., p. 150.

          3 Ibid., p. 149.


seems to produce sentences that are one class by form and another by

meaning. Moule thinks that this is a problem:

          The difficulty of classifying is illustrated by sentences which

      belong by meaning in one class, but by form in another; e.g.:

      (1) in form, (2) in meaning: II Tim. ii:12 ei] u[pome<nomen, kai>

      sunbasileu<somen; II John 10 ei@ tij e@rxetai pro>j u[ma?j kai> tau<thn

      th>n didaxh>n ou] fe<rei, mh> lamba<nete au]to>n. Both these might well

      have had ean with Subj. in the protasis. (1), in meaning, (2) in

      form: I Cor. ix. 16 e]a>n ga>r eu]aggeli<zwmai, ou]k e@stin moi

     kau<xhma . . . .1

          But is this really a problem? The two examples listed as

being class 1 (possible or actual) by form and class 2 (recurrent or

future conditons) by meaning, 2 Tim. 2:12 and 2 John 10, are straight-

forward first class conditions in both form and meaning. It is futile

to suggest what "might well" have been written, for the text has been

set down as God wanted it given. One might feel that an exegetical

problem could be solved by treating 2 Tim. 2:12 as a hypothetical

condition, but such is not the case. It is a first class and needs

to be interpreted as such.  So with 1 Cor. 9:16; it is presented as

a hypothetical condition, even though facts outside the condition

establish that it is an actual situation. There is no need to rewrite

the statement.

          It seems reasonable to ask, If one's analysis produces such

apparent contradictions as Moule felt his did, should the analysis be

pursued? It seems-unlikely that such diverse situations would arise in

the normal development of a language. Significant also is the fact that

no other grammarian has followed Moule in this approach. All seem to


          1 Moule, Idiom Book, p. 149.


have recognized the weakness in his system.


Friedrich Blass - Albert Debrunner

          In 1911 Friedrich Blass published his Grammar of New Testament

Greek. It passed through several editions and translations and was

continued after his death by Albert Debrunner, a professor of Indo-

European and classical Philology at the University of Bern.  Robert

W. Funk of the University of Montana prepared a new translation and

revision of this work as a companion grammar to Arnt and Gingrich's

Lexicon. Funk's revision is the one under consideration.


Statement of His Position

          Blass-Debrunner recognizes five forms of conditional sentences

in Classical Greek, four of which are present in the New Testament:

          (1) Ei] with the indicative of all tenses denotes a simple

      conditional assumption with emphasis on the reality of the assump-

      tion (not of what is being assumed): the condition is considered

      'a real case.'

          (2) Ei] with the optative presents something as thought of,

     without regard for reality or unreality, and emphasizes the

     hypothetical character of the assumption: 'a potential case.'

          (3) Ei] with an augmented tense of the indicative marks the

      assumption as contrary to fact: 'an unreal case.'

          (4)   ]Ea<n with the subjunctive denotes that which under certain

     circumstances is expected from an existing general or concrete

     standpoint in the present: 'case of expectation.'

          (5) Ei] with the optative also specifies repetition in past


          Of these five forms, (2) has almost disappeared from the New Testament

and (5) has completely disappeared. Blass-Debrunner thus comes close

to Robertson's (and Gildersleeve's) four conditions. Indeed, Funk


          1 Blass-Debrunner, Grammar, p. 188.


speaks of Robertson's organization as "especially lucid."1 Unfortunately,

neither Blass-Debrunner's nor Funk's comments have the same quality, so

some explanatory comments are necessary.

          The first group, the "real case," speaks of

     . . . a present reality = 'if . . . really' (as you say, as is

     believed, as you see, etc.) or = 'if therefore' (resulting from

     what has been said), often closely bordering on causal

       'since' . . . .2

Were one to change the adverb "often" to "infrequently," he would be

closer to the truth of the New Testament. He comes closer in his

first statement that this condition places emphasis on "the reality of

the assumption (not of what is being assumed)." This distinction between

the statement of a situation and the actual situation itself must be

maintained. In Matthew 12:27, for example, Jesus speaks of His

exorcism in a way that is directly counter to the reality of the situa-

tion. This fits Blass-Debrunner's statement.

          The third group, ei] with an augmented indicative verb,

indicates a condition that is contrary to fact. Unlike Classical Greek,

the "addition of a@n to the apodosis is no longer obligatory."3  These

conditions are "remarkably scarce in Paul."4


          1 Blass-Debrunner, Grammar, p. 189.

          2 Ibid., p. 189.

          3 Ibid., p. 182.

          4 Ibid., p. 182.


          The fourth group, e]a<n with the subjunctive, refers to future

situations. These situations may be of a general or specific nature

("general or concrete standpoint"), corresponding to the often-mentioned

present general condition. The use of e]a<n with the subjunctive to

refer to future time is normal. "There is," he notes, "no certain

example of e]a<n with the future indicative in the NT."1  This is not

unexpected, for there has always been a close affinity between the

future indicative and the subjunctive. Indeed, the origin of the future

indicative may well have been a mixture of "the Indo-European future,

which denoted future time, and the subjunctive of the sigmatic aorist."2

This close relationship has been noted before.


Evaluation of His Position

          Blass-Debrunner agrees with Robertson in his stress upon the

limitation of the reality of the first class condition to the statement,

not the situation. Although Funk respects Robertson's analysis, he

retains the original terminology of Blass's work. While there may

not be general agreement between this grammar and that of Robertson on

titles, there is on the forms and their significance.


Nigel Turner

          Nigel Turner's role in finishing the grammar started by Moulton

has already been set forth. His volume on syntax represents conditional


          1 Blass-Debrunner, Grammar, p. 190.

          2 Ibid., p. 166.


sentences under their respective parts: moods, tenses, etc. Only

towards the end does he summarize his conclusions.

Statement of His Position

          In outline form his analysis is:

       (i)          ei] with indic., representing the simple assumption . . .

       (ii)         ei] with opt., representing the "potential" conception . . . .

       (iii)        ei] with aor. or impf. indic., representing an assumption as

                    not corresponding with reality . . . .

      (iv)         e]a<n with subj., indicating an expected result based on the

                    present general or particular circumstance . . . .1

Turner presents a system of analysis that combines features of Moulton

along with some of Goodwin. It is instructive to see what new insights

this gives for each of these conditions.


Details of His Position

          Turner does not give a detailed discussion of the first type

of condition except to note that ei] with the future indicative,

unlike e]a<n with the present subjunctive, calls attention to "The

feeling of definiteness and actual realization [that] accompanies it.

It is almost causal."2  He recognizes the problem posed by 2 Timothy

2:12, but offers no suggested answer: "The difficulty about this view

is 2 Ti 212 ei] a]parnhso<meqa, where the condition was surely conceived

as no more than hypothetical."3


          1 Turner, Syntax, p. 319.

          2 Ibid., p. 115.

          3 Ibid., p. 115.


          Turner notes that in the third group of conditional sentences,

the "unreal" conditions, the past tenses are used and retain their

proper Aktionsart. Thus the imperfect stands for "what should be now"

and the aorist for "what should have been."1  In both cases, reality is

not found in what the speaker wants it to be or to have been.

          His discussion of the fourth group of conditions (protasis =

e]a<n with the subjunctive) includes the concept of particular and

general as defined by the Aktionsart (as he understands it) of the

verbs employed:

          (1) Present: very common in Koine. In a general and iterative

     sense, as "condicio universalis" . . . , the pres. subj. denotes a

     hypothesis which can occur over and over again (present Aktionsart).

     The most common example of this condition in the Ptol. Pap. is

     stereotyped phrases in decrees and punishments, having a continual


          (2) Aorist: This represents a definite event as occurring

      only once in the future, and conceived as taking place before the

      time of the action of the main verb. It is expectation, but not

      fulfillment as yet.2

Here he evidences the common, though erroneous, view of the significance

of the aorist tense. Further, he does not make the concept of particular

and general a basis of classification, only of interpretation.

          Not all conditional sentences fit neatly into this four-group

package, and Turner speaks of "a liberal mixing in the various categories

of conditional sentences."3 The specifics remain to be explored, but


          1 Turner, Syntax, p. 91.

          2 Ibid., p. 114.

          3 Ibid., p. 319.


it would seem, as with Moule, that any system that produces such

results needs to be reexamined.


H. E. Dana and Julius R. Mantey

          The appearance of Dana and Mantey's Manual Grammar probably

accounted for the disappearance of Robertson's Shorter Grammar of the

Greek New Testament as a standard intermediate text.1  In their book

these authors summarize Robertson's analysis and add some interesting

terminology and explanations of their own.


Statement of Their Position

          The basis upon which they classify conditional sentences is

"the attitude that they express with reference to reality."2  This is

expressed through the mood of the protasis. The indicative points to

a condition from the viewpoint of reality. The subjunctive and

optative moods point to a condition from the viewpoint of probability.3

Those conditions which use the indicative mood are divided into

two sub-classes. The first, termed the "simple condition," presents

"one fact as conditioning another." In this form "nothing is implied

as to whether or not this fact actually exists."4 The second sub-class


          1 Now available as a reprint: A. T. Robertson and W. Hersey Davis,

A New Short Grammar of the Greek Testament, 10th edition (Grand Rapids:

Baker Book House, 1977).

          2 Dana and Mantey, Grammar, p. 287.

          3 Ibid., p. 287.

          4 Ibid., p. 287.


implies that "this fact has not been realized, and therefore does not

exist. This we call the contrary to fact condition."1  These two types,


      . . . assume that the premise is either true or untrue. The speaker

     takes for granted that which he assumes is true, as in the simple

     condition; or that it is known not to be true, as is the case in the

     contrary to fact condition. The indicative, being the mood for

     reality, is regularly used in this type of sentence.2

Using Matthew 12:27 as a test case, it is clear that their first state-

ment (The condition implies nothing about the actual facts of the case.)

is more accurate than the second one. Jesus did take His assumed link

with Beelzebub for granted, but only for the sake of the argument.

          Conditions utilizing the subjunctive mood are termed the "more

probable future conditions," and those with the optative are the "less

probable future conditions."3 All of these conditions utilize the

particle a@n whether by itself or in combination with ei] (ei] + a@n =



Evaluation of Their Position

          Their observations on the construction of the four types of

conditional sentences are identical to Robertson's. They identify three

types of irregular forms of conditional sentences: (1) mixed conditions,


          1 Dana and Mantey, Grammar, p. 287.

          2 Ibid., p. 288.

          3 Ibid., p. 287.

          4 Ibid., p. 288.


(2) implied conditions "in which the apodosis is expressed and the protasis

implied in a participle (I Tim. 4:4), imperative (Mk. 1:17), or question

(Mt. 26:15)," and elliptical conditions.1

          Although their terminology is somewhat different than that of

Robertson, the closeness of their position to his places them under

the same evaluation.


Robert W. Funk

          As Professor of Religious Studies at the University of Montana,

Robert W. Funk has not only translated and enlarged the standard

grammar of Blass-Debrunner, but has produced one of his own. His

discussion of conditional sentences reflects the basic system of

Robertson, but with significant comparisons to Classical grammars,

especially the work of Smyth.


Statement of His Position

          The first class condition presents "a simple conditional

assumption with emphasis on the reality of the assumption (but not on

the reality of what is being assumed)."2 This is the same statement he

used in his revision of Blass-Debrunner, and the same clarification is

necessary. By "assumption" he means the statement of the situation,

and by "what is being assumed" he means the situation itself. The use

of the future tense in both the protasis and apodosis is the equivalent


          1 Dana and Mantey, Grammar, p. 291.

          2 Funk, Grammar, II, p. 680.


of Smyth's "future most vivid" condition, but Funk recognizes it as a

special form of the first class condition.1

          His analysis of the second class is the same as Robertson. The

third class of conditions utilizes the subjunctive mood in the protasis

and is called "a probable case."2  The use of the present tense in the

apodosis "often gives the condition a generalizing force (indicated

by event in the translation)."3  This is what Smyth termed the "future

more vivid" condition.


Evaluation of His Position

          Funk demonstrates the advantages of classification by reality

as expressed in the mood of the protasis. He is able to present a

simple, easily comprehended, system and account for the many subspecies

recognized by other grammarians. His improvement upon Robertson's

system is more in terms of organization than content.


William Sanford La Sor

          William Sanford La Sor is more familiar to scholars for his work

in Hebrew rather than Greek, for he is Professor of Old Testament at

Fuller Theological Seminary. He has, however, published a two-volume

handbook of Greek grammar based on an inductive study of Acts. In this

work he follows a system unique to modern Koine grammarians: that of


          1 Funk, Grammar, II, p. 684.

          2 Ibid., p. 683.

          3 Ibid., p. 683.




Statement of His Position

          The outline of his position is almost identical to that of


     Simple Condition,

          Protasis = ei] + indicative

          Apodosis = indicative or equivalent

      Unreal Condition

          Protasis = ei] + past indicative

          Apodosis = (a@n) + past indicative

      Present General Condition

          Protasis = e]a<n + subjunctive

          Apodosis = present indicative

      More Vivid Future Condition

          Protasis = e]a<n + subjunctive/other

          Apodosis = future indicative

      Less Vivid Future Condition

          Protasis = ei] + optative

          Apodosis = a@n + optativel

          This classification is based upon time and reality, as was

that of Goodwin. La Sor states that:

     Present conditions can be only noncommittal or general . .

     Past conditions can be noncommittal or contrary to fact . . . .

     Future conditions can be only probable. But the degree of probability

     in the speaker's mind is variable. There is a more probable (or

     "more vivid") future condition. . . and a less probable (or "less

     vivid") future condition . . . .2

          He offers some interesting comments on the various types of

conditions. For example, he says that the simple condition [obviously]


          1 William Sanford La Sor, Handbook of New Testament Greek, 2

volumes (Grand Rapids: Wm. B. Eerdmans Publishing Company, 1973), II,

p.. 225.

          2 Ibid., p. 222.


may refer to past time by using a past tense in the protasis. This

form is very similar to that of the unreal condition, which also uses

a past (augmented) tense in the protasis. Further, he notes that the

regular negative particle for the protases of past simple conditions

is mh<, even with the indicative tense: "Where ou] is used, it probably

negates a word in the protasis and not the entire protasis."1 He also

identifies mh< as the negative particle of unreal conditions, "even though

it [the verb] is indicative."2


Evaluation of His Position

          La Sor evidences the same problem that Goodwin does in his

analysis: classification more by interpretation than by form. He notes,

for example, that the present general condition, which uses e]a<n and the

subjunctive mood, is often "close to, if not identical with the simple

condition . . . except for the use of         e]a<n + sbjtv. [sic]"3

          The more vivid future condition "is expressed by using say in

the protasis, generally with the subjunctive, and a future indicative or

equivalent in the apodosis."4  La Sor then remarks that "there is

considerable variation in the protasis of this type of condition: and

offers these examples:


          1 La Sor, Grammar, II, p. 223.

          2 Ibid., p. 224.

          3 Ibid., p. 224.

          4  Ibid., p. 224.


     With ei] + subjtv: Lk. 9:13; I Cor. 14:5.

     With ei] or e]a<n + fut. ind.: Ac. 8:31; 2 Tim. 2:12.

     With ei] + pres. ind.: Matt. 8:31; I Cor. 10:27.1

Again, it would seem that interpretation more than form is the guiding

principle of classification. The specific verses will be analyzed in

the following sections of this study. La Sor's position will have to

be carefully evaluated, for he represents a unique position among

modern Koine grammarians.


                              Summary of Koine Grammarians

          The twelve Koine grammarians discussed in this section may be

classified into two groups: (1) those who do not follow Robertson: Moule

and La Sor, and (2) those who do: all the rest. The first group pursues

either their own approach, such as Moule, or follow most Classical

grammarians, as La Sor does. They are a decided minority among Koine

grammarians, both early and modern.

          The second group follows the system presented by Robertson in

Koine studies and Gildersleeve in Classical grammar. These classify

conditional sentences according to the reality of the condition as

expressed by the mood of the protasis. The first group basically follows

the time of the condition as the principle of classification.

          The relation of the condition to reality is either actual or

potential. The moods utilized in the protasis indicate these relation-

ships: indicative = actual, subjunctive or optative = potential. The


          1 La Sor, Grammar, II, p. 224.


first group of conditions, those using the indicative mood, may be real

or unreal. The speaker may assume that the condition is a real condition,

or he may assume that it is not. In either case the indicative mood is

used, and in both cases the reality is limited to the statement, not

the situation. The second group, those conditions presented as

potential, involve two degrees of potentiality. Here the grammarians

struggle with terminology. Dana and Mantey are the least ambiguous with

their terms "more probable" and "less probable." The degrees of

probability are indicated by the subjunctive and optative moods,






                                        CHAPTER II


                              THE SIMPLE CONDITION


          A survey of the various grammarians and their attempts to

classify conditional sentences is helpful, but not determinative. Only

an inductive analysis of the New Testament examples can offer definitive

evidence for one system over another. Such an inductive study is the

goal of the next four chapters.



          The following steps were taken to achieve this goal: (1) All

conditional sentences were identified and entered on index cards. This

was accomplished by looking at every reference containing a conditional

particle as listed in Englishman’s Greek Concordance. Each card showed

the protasis, apodosis and verb parsings of every condition. Supplied

verbs were entered based upon context and reference to the New American

Standard Bible.

          (2) This information was then entered on punch cards for ease

in mechanical sorting. The cards were then sorted into the various groups

as indicated in the following chapters. This arrangement also facilitated

rapid cross-checking of the various types of conditions.

          (3) The information on each punch card was again verified from

the Greek New Testament and the lists were prepared. There are no

complete listings of all conditional sentences in the New Testament.


Robertson offers a full, albeit incomplete listing in the Appendix of his

grammar,1 and J. W. Roberts has even more extensive lists in his disser-

tation.2  After the author had completed his lists, he compared them

to Robertson and Roberts and made the necessary adjustments. Since

Roberts' were more complete, they were usually consulted. Careful

study showed that while they were extensive, they were not perfect. The

results, it is hoped, will be even more extensive.

          It is readily acknowledged that these lists are not perfect.

Many apodosis verbs have to be supplied from the context, and honest

differences of opinion exist as to what tense and mood is to be

inserted. Also, in a study of this magnitude, omissions and errors

are possible, although every effort has been made to keep such to a

minimum. Individual questions about specific conditions, though, will

not affect the general picture that emerges from the data, and this is

the desired goal.

          In order to avoid problems with certain incorrect implications

drawn from A. T. Robertson's terminology and its inadequacies, the

"neutral" terminology of Dana and Mantey will be employed. Again, their

definition of a simple condition:


          1 A. T. Robertson, A Grammar of the Greek New Testament in the

Light of Historical Research Nashville, Tennessee: Broadman Press, 1934),

pp. 1416-1419.

          2 J. W. Roberts, "The Use of Conditional Sentences in the Greek

New Testament as Compared with Homeric, Classical and Helenistic Uses,"

unpublished Ph.D. dissertation, the University of Texas, 1955.


     This condition was used when one wished to assume or to seem to

     assume the reality of his premise. Ei] occurs regularly in the

     protasis, with any tense of the indicative. There is no fixed

     form for the apodosis--any mood or tense may occur.1

          Appendix I lists the occurrences of the simple condition, and

verifies their observation: most moods and tenses do appear. The summary

of these occurrences is:

          Protasis using ei] with the present indicative . .           221 examples

          Protasis using ei] with the aorist indicative. . . 56 examples

          Protasis using ei] with the future indicative. . . 22 examples

          Protasis using ei] with the perfect indicative . .           12 examples

                                                                                311 examples


                           The Conditional Particle Ei]

          The discussion of the simple condition begins with a brief

analysis of its components. Taking them in order of appearance, the

first subject is the condition particle ei].


                                   The Significance of Ei]

          When used by itself, ei] may be several things, including

(1) a conditional particle, (2) a type of aposiopesis (a sudden breaking

off of what is stated), especially as a replacement for the Hebrew Mxi  

and (3) an interrogative particle.2


          1 H. E. Dana and Julius R. Mantey, A Manual Grammar of the Greek

New Testament (New York: The MacMillan Company, 1966), p. 289.

          2 William F. Arndt and F. Wilbur Gingrich, A Greek-English

Lexicon of the New Testament and Other Early Christian Literature,

second edition (Chicago: University of Chicago Press, 1979), pp. 219-



As a Conditional Particle

          Ei] regularly appears with the indicative mood to indicate the

simple condition. It also occurs with the subjunctive mood in Luke 9:13;

Romans 11:14; I Corinthians 9:11 (T.R.), 14:15; Philippians 3:11,12;

I Thessalonians 5:10; and Revelation 11:5. Since the mood of the

verb determines the type of condition, not the particle employed,

these specific passages are discussed in the next chapter.   Ei] also

appears with the optative in the less probable future conditions.


As an Interrogative Particle

          Not all occurrences of ei] mark conditional sentences, for it is

used to indicate questions, especially indirect ones. Robertson notes

that its use with direct questions is close to an elliptical condition

and suggests Mark 15:44 and Luke 23:6 as examples.1 This is parallel

with its use in marking out direct quotes, frequently serving as

quote marks in English (e.g. Matthew 12:10). Robertson further

suggests that this use may be due to the fact that the Septuagint

utilizes ei] at times to translate the h-interrogative, as here in

Matthew 12:10.2  This usage is a change from the usual Classical use of

ei]. One cannot automatically think "conditional sentence" whenever

he observes this particle in the text.


          1 A. T. Robertson, Grammar, p. 916.

          2 Ibid., p. 916.


In Aposiopesis

          A third important use of ei] is in aposiopesis,

      . . . the sudden breaking off of what is being said (or written),

     so that the mind may be more impressed by what is too wonderful,

     or solemn, or awful for words: or when a thing may be, as we some-

     times say, "better imagined than described."1

In this construction, the protasis is stated and the writer drops the

sentence, letting the reader draw his conclusion, as in Luke 19:42.

A related use is to translate the Hebrew particle Mxi. This is a

recognized Hebraism and is not a conditional sentence.

          Ei] does have a number of uses, but the most important one is

that of the conditional particle. Before pursuing its most common

use, brief mention should be made of its appearances in combination with

other particles.


                      The Significance of Ei] with Other Particles

Ei] a]ra<

          There are only two New Testament examples of ei] a@ra: Mark 11:13

and Acts 8:22. This combination emphasizes the assumption: "When placed

after pronouns and interrogative particles, it refers to a preceding

assertion or fact, or even to something existing only in the

mind . . . ."2  Acts 8:22 (deh<qhti tou? kuri<ou ei] a@ra a]feqh<setai< soi)


          1 E. W. Bullinger, Figures of Speech Used in the Bible (Grand

Rapids: Baker Book House, 1968 reprint of 1898 edition), p. 154.

          2 Joseph Henry Thayer, Thayer's Greek-English Lexicon of the New

Testament, reprint of Corrected Edition (Marshallton, Delaware: The

National Foundation for Christian Education, n.d.), p. 71.


is translated by Thayer as "If, since thy sin is so grievous, perhaps

the thought etc."1  This usage is similar to ei] pw?j, as will be seen




          Six undisputed examples of ei]pe<r are found in the New Testament:

Romans 8:9,17; I Corinthians 8:5, 15:15; 2 Thessalonians 1:6 and

I Peter 2:3. I  2 Corinthians 5:3 B, D. E. and G support ei]pe<r.  Ei] ge< is

the accepted reading, supported by x, C, K, L and P.2  This combination

emphasizes the concept under discussion: the particle per means "to do

a thing to the limit (beyond), thoroughly."3  Suggested translations of

ei]pe<r include "if indeed, if after all, since."4  Use of this particle

combination emphasizes the veracity of the condition being discussed.

Interestingly, only Paul employs this construction.


Ei] ge<

          The combination of ei] ge< is represented by five New Testament

examples: 2 Corinthians 5:3; Galatians 3:4; Ephesians 3:2, 4:21; and

Colossians 1:23. This combination emphasizes the conditional nature of


          1 Thayer, Lexicon, p. 71.

          2 W. Robertson Nicoll, editor, The Expositor’s Greek Testament,

5 volumes (Grand Rapids: Wm. B. Eerdmans Publishing Co., 1967), II, p. 66.

          3 Robertson, Grammar, p. 1154.

          4 Arndt and Gingrich, Lexicon, p. 219.


the thought. Commenting on ei] ge< kai> ei]kh? in Galatians 3:4 Lightfoot


      Ei@ ge< leaves a loophole for doubt, and kai< widens this, implying

      an unwillingness to believe on the part of the speaker. Hermann's

     distinction . . . that ei@ ge< assumes the truth of a proposition

     while ei@per leaves it doubtful, requires modifying before it is

     applied to the New Testament, where ei@per is, if anything, more

     directly affirmative than ei] ge<.1

          Arndt and Gingrich suggest "if indeed, inasmuch as" as possible

translations of this combination and render Galatians 3:4 as "have you

experienced so many things in vain? If it really was in vain . . . .”2

The particle ge< performs its usual emphatic function, strengthening the

word to which it is attached, and stresses the conditional nature of

the concept.


Ei] kai<

          The combination ei] kai< is used to introduce concessive clauses,

and is usually translated "even if."  Robertson represents the majority

opinion when he writes that concessive clauses "are really just

 conditional clauses with the addition of kai<."3  Blass-Debrunner agrees

with this evaluation.4  Burton, as Robertson noted, draws a major

distinction between conditional and concessive clauses:


          1 J. B. Lightfoot, The Epistle of St. Paul to the Galatians

(Grand Rapids: Zondervan Publishing House, 1966 reprint of 1866 edition),

pp. 135-136.

          2 Arndt and Gingrich, Lexicon, p. 152.

          3 Robertson, Grammar, p. 1026.

          4 Friedrich Blass and Albert Debrunner, A Greek Grammar of the

New Testament and Other Early Christian Literature, translated and revised

by Robert W. Funk (Chicago: University of Chicago Press, 1961), p. 190.


          The force of a concessive sentence is thus very different

      from that of a conditional sentence. The latter represents the

      fulfillment of the apodosis as conditioned on the fulfillment of

      the protasis; the former represents the apodosis as fulfilled in

      spite of the fulfillment of the protasis.1

          He does recognize, though, that there are times when the two

clauses become almost identical:

     Yet there are cases in which by the weakening of the character-

     istic force of each construction, or by the complexity of the

     elements expressed by the protasis, the two usages approach so

     near to each other as to make distinction between them difficult.2

          Few Koine grammarians follow Burton in this distinction. For

purposes of classification, concessive clauses were omitted from the

lists of conditional sentences in the appendices. Including them would,

of course, alter the total number of examples, but would not affect

the general conclusions.


Ei] pw?j

          The last particle combination is ei] pw?j.  This occurs three

times with the indicative mood (Romans 1:10, 11:14 and Philippians

3:11), twice with the subjunctive (Romans 11:14 and Philippians 3:11),

and once with the optative mood (Acts 27:12). This combination, like

ei] a@ra, serves to heighten the question involved. It is uniformly

translated in the Authorized Version as "if by any means."

          While each of these particle combinations has its special point

of emphasis, it is the conditional use of ei] that is the center of

attention. The next area of discussion will be that of the negative


          1 Ernest DeWitt Burton, Syntax of Moods and Tenses in New

Testament Greek (Chicago: University of Chicago Press, 1897), p. 112.

          2 Ibid., p. 112.




particles employed in the simple condition.


                      Negative Particles in Simple Conditions

          The topic of negative particles in conditional sentences shows

how far Koine Greek has moved from Classical Greek, for the divergence

in form and style is marked.


The Classical Pattern

          W. W. Goodwin will serve as the standard for presenting the

classical pattern of negative particles in conditional sentences:

          The negative particle of the protasis is regularly mh<, that

of the apodosis is ou].

          When ou] is found in a protasis, it is generally closely connected

with a particular word (especially the verb), with which it forms a

single negative expression; so that its negative force does not (like

that of mh<) affect the protasis as a whole.1

Though there are, of course, exceptions, Goodwin's rule states the

general case for the Classical literature. The Koine situation is

quite different.


The Koine Pattern

          Statement.--Alexander Buttmann, an early Koine grammarian,

recognized the distinction between Koine and Classical Greek on this


          The use of ou] in the protasis of a conditional sentence occurs

      in the N.T. relatively very often; so that we are


          1 W. W. Goodwin, Syntax of the Moods and Tenses of the Greek

Verb (Boxton: Ginn and Company, 1893), p. 138.



      justified in inferring a difference in usage, since in classical

     writers this use is only exceptional.1

          A later writer, J. H. Moulton, approvingly quotes Blass's rule for

the negative particles in Koine: "'All instances,' he says, 'may

practially he brought under the single rule, that ou] negatives the

indicative, mh< the other moods, including the infinitive and


          Robertson agrees, noting that "The negative of the protasis in

the first class condition is practically always ou] in the N.T. We have

ei] ou] as a rule, not ei] mh<."3  This rule is not absolute, and he

identifies five exceptions, listed in the next section. Both he and

Moulton see the simple absolute rule of ou] with the indicative and mh  

with the other moods as a goal "not yet reached in the N.T." but almost

completely met in Modern Greek.4


          Specific examples.--The authorities list different totals for

the various combinations. Moulton finds thirty-one examples of ei] ou]  

in simple condition protases,5 Robertson thirty-four,6 and Roberts


          1 Alexander Buttmann, A Grammar of the New Testament Greek, trans-

lated and edited by J. H. Thayer (Andover: Warren F. Draper, Publisher,

1873), pp. 344-345.

          2 J. H. Moulton, Grammar of New Testament Greek. Volume I:

Prolegomena, third edition (Edinburgh: T. & T. Clark, 1908), p. 170.

          3 Robertson, Grammar, p. 1101.

          4 Ibid., p. 1101, Moulton, Prolegomena, p. 170.

          5 Moulton, Prolegomena, p. 171.

          6 Robertson, Grammar, p. 1111.



thirty-five.1  Whatever the discrepancies, the preponderance of this form

is clear when compared to only five examples of ei] mh<:  Mark 6:5,

I Corinthians 15:2, 2 Corinthians 13:5, Galatians 1:7 and I Timothy 6:3.

Of these five, only Mark 6:5 and I Timothy 6:3 have unanimous acceptance

as examples of the classical pattern.

          Godet considers I Corinthians 15:2 "a pleonasm arising from the

mixing of the two following constructions: excepting if (e]kto<j ei])

and: if not (ei] mh<).2  Lenski agrees with this identification.3

          Lightfoot explains Galatians 1:7 so:

     Ei] mh< seems always to retain, at least in this stage of the

     language, its proper exceptive sense, and is not simply oppositive,

     though it frequently approaches nearly to a]lla< . . . .4

This construction "may either state an exception to the preceding

negative clause (= except, save) or merely qualify it (= but only),

as it does in Luke iv. 26 . . . and in Gal. 1. 7 . . . ."5 The same

basic construction is found in 2 Corinthians 13:5 – ei] mh<ti a]doki<moi e@ste.

          The ratio of five to thirty-five examples gives credence to

Buttmann's claim that the Koine shows a wide divergence from the

Classical usage in the negative particles in the simple condition. The

presence of a relatively few examples of mh< show that the Classical

heritage has not been completely lost in the New Testament.


          1 Roberts, "Conditional Sentences," p. 153.

          2 Frederick Louis Godet, Corinthians, II, p. 269.

          3 Lenski, Corinthians, p. 629.

          4 Lightfoot, Galatians, p. 77.

          5 Nicoll, Testament, III, p. 156.


                              ]Ea<n with the Indicative Mood

          Although, the general principle of simple conditions is that the

protasis consists of ei] with the indicative mood, there are exceptions.

A few such conditions employ ean with the indicative in an apparent

contradiction to this principle: Mark 8:3; Luke 19:40; John 8:54,

21:22,23; Acts 8:31; I Thessalonians 3:8; I John 5:15 and Revelation



                               Explanation of the Form

          While these nine exceptions are a small percentage of the total,

they are a phenomenon that needs explanation. Several explanations may

be offered.


Development of the Language

          First of all, Koine Greek represents a stage in the historical

development of the language. As the language progressed from Classical

through Koine to Modern, the distinction between ei] and e]a<n faded,

especially in respect to their respective moods. In the later Koine

"the use of e]a<n with the ind. is rather more frequent . .        . Finally

ei] came to be 'a mere literary alternative.'"1  Blass-Debrunner also

recognizes this consequence of linguistic development.2   The New

Testament, therefore, represents a period in which the general principle

is operative, but a transition is evident. Robertson summarizes: "In

general, the difference between ei] and e]a<n is considerably lessened


          1 Robertson, Grammar, p. 1009.

          2 Blass-Debrunner, Grammar, p. 190.


in the koinh<, though it must be remembered that e]a<n was never confined

to the subj. nor ei] to the ind. and opt."1


Textual Emendations

          While the development of the language may be one consideration

in explaining these apparent exceptions, textual emendations are

another. Buttmann raises this observation and caution:

          It is, indeed, not to be denied that the instances in question

     almost disappear amid the multitude of those that are grammatically

     regular, and suspicion may also be raised by the circumstances

     that hardly a single passage with the Indicative is completely

     beyond question critically. Yet when we consider that in countless

     passages with the Subjunctive not the smallest variation is found

     (which would not be the case if the Indicative were chargeable solely

     to the copyists), it is far more probable that, where a diversity

     of readings occurs in such a number of instances, this fact results

     from the circumstance that the copyists, commentators, etc., early

     altered the Indicative which gave them offence.2

          Winer also notes the significance of textual variations in

his comments on the subject.3  Both grammarians, however, recognize the

legitimate identification of e]a<n with the indicative in the New Testa-

ment in spite of the fact that most of the examples show textual


          How wide-spread is this textual difficulty? A review of the

available evidence supports the claim of Buttmann: such constructions


          1 Robertson, Grammar, pp. 1009-1010.

          2 Buttmann, Grammar, p. 222.

          3 G. B. Winer, A Grammar of the Idiom of the New Testament,

seventh edition, revised by Gottfried Lunemann, translated by J. H.

Thayer (Andover: Warren F. Draper, Publisher, 1893), p. 294.


are part of the original text. Robertson discusses several passages which

have been challenged,1 and Roberts lists the textual evidence for

some of them and adds a few examples of his own.2

          Thus two things seem to be born out: (1) there has been some

alteration of the text in apparent attempts to bring specific passages

into conformity with the general rule and (2) e]a<n was used with the

indicative mood in some passages of the original text of the New



                                  Significance of the Form

          If the use of e]a<n with the indicative is part of the original

text, then what is its significance? Specific answers vary: Robertson,

of course, sees no special significance. The key for him is the mood,

not the particle. Therefore he sees no basic difference between the

two types of protases, for each uses the indicative mood.4  Both

represent simple conditions. Burton agrees, though he speaks less


          In a few instances say is used with the Present Indicative in

      the protasis of a conditional sentence, apparently to express a

      simple present supposition. I Thess. 3:8, I John 5:15.5


          1 Robertson, Grammar, p. 1010.

          2 Roberts, "Conditional Sentences," pp. 146-149.

          3 Moulton, Prolegomena, p. 168.

          4 Robertson, Grammar, p. 1007.

          5 Burton, Syntax, p. 103.


      Blass-Debrunner offer no comment either way.1

          Roberts approaches the subject from the historical side and

discusses the use of e]a<n in conditional sentences in the Septuagint.

In this section of his dissertation he relies upon the work of James

Sterenberg who wrote a dissertation for the University of Munich in

1908 entitled "The Use of Conditional Sentences in the Alexandrian

Version of the Pentateuch."  This work was not available to the author,

so Roberts' conclusions will have to be accepted at face value:

          It will be remembered that Sterenberg noted that the construction

      is used in the LXX (where it occurs with the perfect, present,

     imperfect, and the aorist indicatives) mostly in laws in the protases

     of which transgressions and the like are minutely defined and that it

     is used to render the original thought more exactly to avoid

     ambiguity; e.g., where the verb in the protasis is thought to

     precede in time the event or the immediately preceding verb, or in

     one verb when the event may be supposed as a possible event,

     requiring the indicative, but where both verbs are governed by e]a<n.2

          This, though, is the sense of the construction in the Septuagint.

New Testament examples are not, as Roberts notes, "so related to laws."3

It would seem that the observations based upon the Septuagint do not

fit the New Testament examples.

          Neither do the commentators offer much help. Lenski identifies

the ean of I Thessalonians 3:8 as "looking to the future."4  Lightfoot,


          1 Blass-Debrunner, Grammar, p. 190.

          2 Roberts, "Conditional Sentences," p. 149.

          3 Ibid., p. 149.

          4 R. C. H. Lenski, The Interpretation of St. Paul’s Epistles to

the Corinthians to the Thessalonians, to Timothy, to Titus and to

Philemon (Minneapolis: Augsburg Publishing House, 1961), p. 291.


after defending the indicative in the same verse, equates doubt with

the particle:

          St. Paul speaks with some hesitation here 'if so be ye stand

      fast.' Their faith was not complete (ver. 10). There was enough

      in the fact that they had been so recently converted, enough in

      the turn which their thoughts had recently taken, absorbed so

      entirely in the contemplation of the future state, to make the

     Apostle alarmed lest their faith should prove only impulsive and


The Expositor's Greek Testament notes that the future indicative in

Revelation 2:22 "expresses rather more probability than the subj. with

e]a<n mh<.”2

          What, then, can be said about a distinction in meaning between

e]a<n with the indicative and ei] with the indicative?  Nothing, really.

There are too few examples upon which to build rules, and the distinction,

if any, is nebulous at best. Roberts accurately states the case: "This

writer is able to discover no distinction which can be put in a rule."3

Neither is this one.


                            Significance of Moods and Tenses

          Since the deciding factor in identifying a simple condition

is the indicative mood in the protasis and not the conditional particle,

the next step in discussing the simple condition is to evaluate the

significance of the various moods and tenses in the protasis and apodosis.


          1 J. B. Lightfoot, Notes on the Epistles of St. Paul (Grand Rapids:

Zondervan Publishing House, 1957 reprint of 1895 edition), p. 46.

          2 Nicoll, Testament, V, p. 361.

          3 Roberts, "Conditional Sentences," p. 150.


                           Moods and Tenses in the Protasis

General Observations

          The simple condition is, by definition, limited to the

indicative mood, for it assumes the reality of the condition. The

particle e]a<n does occur, but the key is the mood. A few examples of ei]  

with the subjunctive mood are found in the New Testament, but these are

identified as more probable conditions and discussed under that

heading. The simple condition is limited to the indicative mood in the


          This mood may be used with any tense. The specific data for

New Testament tense usage is listed below:

                    Present tense                     = 69.0%

                    Aorist tense            = 17.5%

                    Future tense            = 6.9%

                    Perfect tense                     = 3.8%

Each of these tenses brings with it the usual verbal significance as

described in the grammars. Which one is used in a given condition

depends upon the action involved in the protasis.

          While there is no rule governing the type of tense used in any

given protasis except the desire of the author, it is clear that the

present tense is used far more than all others combined.


Future Indicative

          One particular combination calls for specific discussion: ei]

with the future indicative. Two things should be noted. First is its


use in Classical Greek, especially by the poets, in threatening or

warning (minatory or monitory) statements. Second is its use in

emphatic assertions or oaths.


Minatory or Monitory Use

          Gildersleeve has called attention to this special use of the

simple condition. Working within the framework of a comparison between

ei] and the future indicative and e]a<n with the subjunctive, he first

notes that the latter is by far more frequent. The distinction between

the two is seen

      . . . whenever it is important to distinguish continued from

     concentrated action, whenever it is important to distinguish over-

     lapping from priority, e]a<n with the subjunctive is preferred.

He continues,

     Now the neglect of this distinction in ei] with the future

     indicative shows a certain coldness, a certain indifference; and

     this added to the general rigor of the logical condition, which

     faces fact in all its grimness, gives a stern, minatory, prophetic

     tone to the future indicative, which commentators and grammarians

      have noticed, but noticed only in passing . . . .2

          Gildersleeve then proceeds to illustrate this from the tragic

poets, showing that the device is frequently employed by them in such


          The New Testament, though, stands in contrast to the tragic

poets of Classical Greek, for there is only one example of this


          1 B. L. Gildersleeve, "On ei] with the Future Indicative and ean  

with the Subjunctive in the Tragic Poets," Transactions of the American

Philological Association, 7:1 (January, 1876), p. 9.

          2 Ibid., p. 9.


construction in its pages.

      Of the New Testament instances of ei] followed by a Future (about

      twenty in number), one, 2 Tim. 2:12, illustrates the minatory or

      monitory force attributed to such clauses by [Gildersleeve].1

          The phrase under discussion, ei] a]rnhso<meqa, ka]kei?noj a]rnh<setai,

is the third line of what may be four lines of an early Christian hymn.2  

The fact is stated "in all its grimness" that "If we shall deny Him,

He also will deny us." Certainly this must be understood as a stern

warning, for eternity seems to hang in the balance. But, it should be

asked, a warning of what to whom?

          Commentators are, as usual, divided on the passage. Some seem

to imply loss of salvation resulting from a believer's turning against

his Lord and denying Him. Hendriksen's words seem to allow for this:

     When a person, because of unwillingness to suffer hardship for

     the sake of Christ and his cause, disowns the Lord ("I do not know

     the man!"), then, unless he repents, he will be disowned by the

     Lord in the great day of judgment ("I do not know you.").3

          Kent understands Paul to be speaking of professed believers in

general and, assuming that some may prove their true colors by denying

Him, warns them of their fate.4  Hiebert agrees with Kent, rejecting

the idea that this denial refers to "a temporary weakness of faith," but


          1 Burton, Moods and Tenses, p. 105.

          2 Homer A. Kent, Jr., The Pastoral Epistles (Chicago: Moody Press,

1958), p. 271.

          3 William Hendriksen, 1 - II Timothy and Titus (Grand Rapids:

Baker Book House, 1965), p. 259.

          4 Kent, Pastoral Epistles, p. 272.


as the conclusion shows, to deny Him "as a permanent fact."1

          Certainly, though, Gildersleeve's identification of this

construction as one presenting grim fact in a coldly logical way aptly

describes Paul's words.


Emphatic Assertions

          The second and more common use of ei] with the future indicative

is in emphatic assertions or oaths. This is especially true in quotes

from the Old Testament where ei] translates the Hebrew particle Mxi.  The

four examples of this are Mark 8:12, Hebrews 3:11, 4:3 and 4:5.

Mark 8:12, ei] doqh<setai th? genea? tau<th shmei?on, is not an

"official" oath like the examples in Hebrews, but it fits the same

pattern:  "No sign shall be given to this generation." As already

indicated, Robertson identifies this as an elliptical condition lacking

the apodosis. Further, he says that this is "really aposiopesis in

imitation of the Hebrew use of im."2  Other grammarians recognize this

construction, including Winer,3 Buttmann,4 and Blass-Debrunner.5


          1 D. Edmund Hiebert, Second Timothy (Chicago: Moody Press, 1958),

pp. 63-64.

          2 Robertson, Word Pictures, IV, p. 331.

          3 Winer, Grammar, p. 500.

          4 Buttmann, Grammar, pp. 358-359.

          5 Blass-Debrunner, Grammar, p. 189.


          The references in Hebrews (3:11, 4:3 and 4:5) are all quotations

from the Septuagint version of Psalm 95:11 – ei] ei]seleu<sontai ei]j th>n

kata<pausi<n mou. The Hebrew version introduces this with the particle

Mxi, whose use in oaths is a normal structure of the language.1  The

Septuagint regularly uses ei] in these situations. The Authorized

Version correctly renders 3:11 as "They shall not enter into my rest,"

while 4:3 and 4:5 are translated, "If they shall enter into my rest."

Modern versions correctly translate all of them the same way: a statement

of warning, "They shall not enter into my rest."

           Thus ei] with the indicative is used in sentences of emphatic

negation, though few of them are found in the New Testament. The

conditional particle is a straight-forward translation of the Hebrew

and is a proper Hebraism.


                         Moods and Tenses in the Apodosis

          There is no specific relationship between the moods and tenses

of the protasis and the apodosis in simple conditions. Robertson

summarizes this point well when he states:

      The apodosis varies very greatly. It all depends upon what one is

      after, whether mere statement, prediction, command, prohibition,

      suggestion, question. Hence the apodosis may be in the indicative

     (any tense) or the subjunctive or the imperative. There is no

      necessary correspondence in tense between protasis and apodosis.

     The variation in the mode of the apodosis has no essential bearing

      on the force of the condition.2


          1 E. Kautzsch, ed., Gesenius' Hebrew Grammar, revised by A. E.

Cowley (London: Oxford University Press, 1910), pp. 471-472.

          2 Robertson, Grammar, p. 1008.


          While there is no essential correspondence between the protasis

and the apodosis, there is a preference for the indicative mood in

the apodosis. The tables in Appendix I yield the following data:

          Apodoses with indicative mood   = 205 or 67.3% of the total.

          Apodoses with imperative mood = 84 or 28.4% of the total.

          Apodoses with subjunctive mood           = 7 or 2.4% of the total.

Again, no rule may be fixed, but the distribution is most reasonable.

Protases which speak in real terms would normally imply apodoses which

also speak in real terms. But language is flexible, and all moods are

possible and do occur.


                            Meaning of the Simple Condition

          Now that the details of amount and construction have been

considered, the way is cleared for a consideration of the basic meaning

of the simple condition. This will cover two areas: (1) particular

and general conditions, and (2) the degree of reality implied by the



                         Particular and General Conditions

          The terminology if not the concept of particular and general

conditions has entered Koine studies through the work of Goodwin, hence

his definitions will be the starting point.


The Position of Goodwin

          Goodwin defines particular and general thusly:

          A particular supposition refers to a definite act or to several

     definite acts, supposed to occur at some definite time (or

     times) . . . .


          A general supposition refers indefinitely to any act or acts

      of a given class which may be supposed to occur or to have occurred

      at any time . . .1

This particular distinction is seen "in all classes of conditions,"

but in some of them it may be distinguished by construction:

          When the apodosis has a verb of present time expressing a

      customary or repeated action, the protasis may refer (in a general

     way) to any act or acts of a given class which may be supposed to

     occur at any time within the period represented in English as


          When the apodosis has a verb of past time expressing a customary

     or repeated action, the protasis may refer (in a general way) to any

     act or acts of a given class which may be supposed to have occurred

     at any time in the past.2

          The key point in the construction of the conditional sentence

is the tense of the apodosis, not the protasis. Goodwin is simply

attributing to the present and imperfect tenses their continual or

repetitive significance.


The Evaluation of Goodwin

          As mentioned earlier, Funk has also noted that "The present

tense in the apodosis often gives the condition a generalizing force

(indicated by ever in the translation)."3  Funk, though, simply notes it

as an observation without making it a rule of classification. In this

he follows Gildersleeve, who notes that all classes of conditional

sentences may be either general or particular, "according to the


          1 W. W. Goodwin, Moods and Tenses, p. 141.

          2 Ibid., pp. 141-142.

          3 Robert W. Funk, A Beginning-Intermediate Grammar of Hellenistic

Greek, 3 volumes (Missoula, Montana: The Society of Biblical Literature,

1973), II, p. 683.


character of the apodosis."1

          Robertson rejects Goodwin's use of this concept as a means of


     This theory calls for "particular" and "general" suppositions as a

     fundamental element. This is a false step in itself. As Gilder-

     sleeve shows, each of the four classes of conditions may be parti-

     cular or general. That point has no bearing on the quality of the


          Does this agree with the facts of the case? Can simple

conditional sentences be particular or general, depending upon the

context as Gildersleeve and Robertson maintain? The answer, of course,

is Yes. A few examples will suffice.

          Romans 4:14 uses two perfect tenses in the apodosis of its

condition and speaks of one past historical event: the giving of the

law – ei] ga>r oi[ e]k no<mou klhrono<moi, keke<nwtai h@ pi<stij.- "for if those

who are of law are heirs, then faith has no value." This fits the

pattern of Goodwin as a particular condition.

          I Corinthians 15:2, however, uses a present tense in the apodosis

and speaks of the specific situation of the Corinthian believers and

their relationship to the gospel – di ] ou$ kai> sw<zesqe, ti<ni lo<gw

eu]hggelisa<mhn u[mi<n ei] kate<xete - "By this (gospel) you are saved, if you

hold fast to the word I preached to you." This does not agree with

Goodwin, for the present tense (sw<zesqe) should make this condition

a general one.


          1 B. L. Gildersleeve, "Studies in Pindaric Syntax," The American

Journal Philology, 3:4 (December, 1882), p. 435.

          2 Robertson, Grammar, pp. 1005-1006.


          Matthew 12:26 – kai> ei] o[ stana?j to>n satana?n e]kba<llei, e]f ]

e[auto<n e]meri<sqh - "And if Satan casts out Satan, he is divided against

himself" - may be taken as a general condition, referring to any time

Satan works against himself. Yet the apodosis uses the aorist e]merisqh,

contrary to what Goodwin would predict.

          Thus the simple condition may be either particular or general

with no special significance attached to the tense of the apodosis.

Funk's point should not be overlooked, though, for the concepts of

particular and general are a legitimate part of interpretation and



                                      Degree of Reality

          One of the most important questions about simple conditions is:

What do they imply about the condition they state? The indicative mood

communicates objective reality, but how is that communicated and

wherein does the reality lie?


Review of the Grammarians

          The various grammarians approach this question in various ways,

but most have arrived at a similar position.


Classical Grammarians

          Goodwin stated his opinion that simple conditions are those which

imply nothing as to the fulfillment [reality] of the condition . .        ,”1

He rejects the idea that language should even have a form which implied


          1 Goodwin, Moods and Tenses, p. 139.


the objective reality of a condition: "The Greek has no form implying

that a condition is or was fulfilled, and it is hardly conceivable that

any language should find such a form necessary or useful."1

          Gildersleeve agrees with Goodwin:

          The Logical Condition [simple condition] states the elements in

      question. It is used of that which can be brought to the standard

      of fact; but that standard may be for or against the truth of the

      postulate. All that the logical condition asserts is the inexorable

      connection of the two members of the sentence.2

His point is that the factuality of the postulate is limited to its

statement, for it is presented as something that can be considered as

fact but not necessarily is fact. The Classical grammarians are in

agreement on this point.


Koine Grammarians

          Early Koine grammarians followed this approach. Green states,

"The conditional particle ei], if with the Indicative in the protasis,

assumes the hypothesis as a fact."3  Winer reflects Gildersleeve's view

by identifying the simple condition as the "Pure Condition."4  Even

Turner, a more recent scholar, uses the general language of "simple



          1 Goodwin, Mood's and Tenses, p. 140.

          2  Gildersleeve, "Pinder," p. 435.

          3 Samuel G. Green, Handbook to the Grammar of the Greek Testament

(London: The Religious Tract Society, [1886]), p. 317.

          4 Winer, Grammar, p . 291.

          5 Turner, Syntax, p. 319.


          More recent grammarians continue this same general approach.

Blass-Debrunner states, somewhat confusingly, "Ei] with the indicative

of all tenses denotes a simple conditional assumption with emphasis on

the reality of the assumption (not of what is being assumed): the

condition is considered 'a real case.'"1  The confusion results from

the distinction between the assumption and "what is being assumed."

The assumption corresponds to the statement, "What is being assumed"

corresponds to the situation. In Matthew 12:27, Jesus offers the

statement with an assumption that it is a real case. He assumes it

to be true. He is not, however, stating the situation as true. The

distinction Blass-Debrunner draws is between these two: the statement

and the situation.

          Dana and Mantey are clearer when they identify the simple

condition as the one that "was used when one wished to assume or to seem

to assume the reality of his premise."2  Funk expands this description:

          In a first class condition the protasis is a simple conditional

     assumption with emphasis on the reality of the assumption (but

     not on the reality of what is being assumed) [reality of statement,

     but not the situation]. It is therefore taken to be a real case

     though it may, in fact, be an unreal case [as in Matthew 12:27].

          Robertson's position is basic to this discussion, since he is

most often quoted in support of one position or another. To show the

development of his thought two statements will be given, one from the


          1 Blass-Debrunner, Grammar, p. 188.

          2 Dana and. Mantey, Grammar, p. 289.

          3 Funk, Grammar, II, p. 680.


first and one from the tenth edition of his Short Grammar:

     The indicative states the condition as a fact. It may or may not

     be true in fact. The condition has nothing to do with that, but

     only with the statement. . . . This condition does assume the

     reality of the condition.1

Then, from the tenth edition:

     This condition assumes the reality of the condition. The indicative

     mode states it as a fact. The condition has nothing whatever to do

     with the actual fact or truth. It is just here that some of the

     grammars have erred in failure to distinguish clearly between the

     statement and the reality. It is the condition taken at its face

     value without any insinuations or implications. The context, of

     course, must determine the actual situation. The indicative mode

     determines only the statement.2

          It is evident in the latter statement that he is trying to avoid

the impression that he considers the simple condition one that affirms

the reality (or actuality) of the situation as an objective fact. It

does affirm the reality of the speaker's assumption or statement, i.e.,

the speaker really assumes (though he may believe otherwise) that the

condition (statement) is true. The statements in his Historical

Grammar, offered in response to Goodwin, are not as clear:

     This condition pointedly implies the fulfilment of the condition.

     It is the condition of actuality, reality, Wirklichkeit, and not

     mere "possibility" as Farrar has it . . . a la Goodwin. This is

     the crux of the whole matter. Once see [sic] that the first class

     condition with the ind. implies the reality of the premise, all

     else follows naturally.3


          1 A. T. Robertson, A Short Grammar of the Greek New Testament

(New York: A. C. Armstrong & Son, 1908), p. 161.

          2 A. T. Robertson and W. Hersey Davis, A New Short Grammar of

the Greek Testament (New York: Harper & Brothers, 1933), p. 350.

          3 Robertson, Grammar, p. 1006.


          It seems, then, that the grammars present a unified view

concerning the degree of reality in the simple condition: It presents

the statement of the condition as true, but without affirming any-

thing about the reality of the actual situation.


Review of the Problem

          The unanimous opinion of the grammarians is not reflected by

some writers and speakers who claim that the simple condition

guarantees the reality of the situation. They suggest the translation

"since," rather than "if." In English this presents the situation

as true, both in the speaker's statement and in objective fact. An

extreme, but by no means isolated example of this has already been

given: "The first class condition implies truth or reality. If . . .

and it is true."1  Sometimes Kenneth Wuest is implicated in this view

on the strength of translation such as Romans 6:5 -

     The Word "if" in the Greek is not the conditional particle of

     an unfulfilled condition. It is a fulfilled condition here, its

     meaning being, "in view of the fact."2

He did not adopt this as a uniform principle of translation, though,

and limited such renderings to places where the context justified them.

This, of course, is at the heart of the situation. There are several

passages where the reality of the statement is in keeping with the

reality of the situation. But many times it is not, e.g. Matthew 12:27.


          1 Jerome Moore, "Four Ways to 'Say 'If,'" The Baptist Bulletin,

45:1 (June, 1979), p. 11.

          2 Kenneth S. Wuest, Treasures From the Greek New Testament

(Grand Rapids: Wm. B. Eerdmans Publishing Company, 1957), p. 89.


One cannot generalize from a context where this is true to all uses of

the simple condition.


          Objections to this view.--Several objections may be raised to

this position. First, a little mature reflection will cause one to be

cautious about agreeing with Moore and others. In his example he

claims, on the basis of a simple condition in Colossians 1:23, that

there is no need to doubt the security of anyone's salvation. This

is not supported from Scripture, for certainly the false teachers of

Acts 20:30 looked and acted like believers before their true colors were

flown. Those in I John 2:19 must have looked like believers before

they went out and demonstrated that they were not. Would Paul have

been so quick to affirm the absolute salvation of all those in the

church where he had not visited (Colossians 1:4, 9)?  Probably not.

          Second, and more significant, the pattern of translating ei]  

by "since" does not hold in all cases. James Boyer shows that Wuest

changes his translation between John 10:35 ("since") and 10:37, 38

("assuming that"), as required by context.1  John Battle has listed all

simple conditions which, in his opinion, present data in the protasis

that are contrary to fact and known to be so by the speaker.2  It is

impossible to use "since" in such verses as I Corinthians 15:13 and


          1 James L. Boyer, "Semantics in Biblical Interpretation,"

Grace Journal, 3:2 (Spring, 1962), p. 33.

          2 John A. Battle, Jr., "The Present Indicative in New Testament

Exegesis," unpublished Th.D. dissertation, Grace Theological Seminary,

1975, pp. 166-168.


still maintain orthodoxy. Battle's list is impressive, and all one has

to do is insert "since" in place of "if" in the verses to realize the

error of this concept. J. Harold Greenlee shows the variety of

situations one encounters in the simple condition:

     Moreover, 2) the speaker may believe that the condition is true:

     "if they have persecuted me (and they have done so) [John 15:20a];

     or 3) he may believe that the condition is not true: "if they have

     kept my word" (but they have not done so) [John 15:20b]; or 4) he

     may be uncertain as to whether it is true: "sir, if you have

     carried him away" (she does not know whether he had or not) [John

     20:15]; or 5) the speaker may even be mistaken in his assumption:

     "if he is sleeping" (but he was in fact dead) [John 11:12].1

          Justification for the translation of e]i by "since," must come

from the context, not the condition. This is an interpretation,

not a translation.

          The correct view.--The simple condition, through the indicative

mood, offers a conditional statement presented as real. It makes no

attempt to speak to the actual situation. If the exegete will maintain

this distinction between statement and situation, then he will not err

in handling this condition. A good summary of this point of view is

found in The Bible Translator:

     When ei] with the indicative is used, it implies that the truth or

     otherwise of the condition is regarded as in principle "determined,"

     i.e. is represented as a fact (although the speaker does not commit

     himself as to whether he believes the condition is true or not.2


          1 J. Harold Greenlee, "'If' in the New Testament," The Bible 

Translation, 13:1 (January, 1962), p. 40.

          2 John Kinje, Jr., "Greek Conditional Sentences," The Bible

Translator, 13:4 (October, 1962), p. 223.


                        Translation of the Simple Condition

          How, then, should the simple condition be translated? The

uniform rule of all grammarians surveyed is to use the English conditional

construction, "if." All major English versions so translate this

condition. Should one want to paraphrase the concept, then the phrase,

"If, as I am assuming . . ." might be employed. The statement of the

simple condition may be in accord with the reality of the situation,

it may be contrary to it, or it may be unknown. All that can be said

is that the speaker is presenting the condition as true in his





                                      CHAPTER III




                                      The Concept

          The simple condition states the matter directly by assuming it

to be true. This, as has been discussed, is an assumption of reality.

But the assumption can also work the opposite way. The speaker can

present the situation and assume it to be false or contrary to fact.

Such a condition is termed the Contrary to Fact condition and is the

subject of this chapter.


                                  The Construction

          These conditions follow a standardized construction:  ei] with

a past (augmented) tense in the indicative mood in the protasis, and

another past (augmented) verb in the apodosis, usually with a@n. This is

as would be expected, for both the simple condition and the contrary to

fact condition are making definite statements; the former assumes them

to be true, the latter false. Robert W. Funk describes them so:

          The assumption is taken to be untrue in the protasis of a second

     class condition [contrary to fact]; it is considered an unreal case

     (whether it is, in fact, is another matter). The conclusion

      follows from the premise.

          The protasis consists of ei] plus a past (augmented) tense of

       the indicative; the apodosis also has a past tense of the indicative,

       usually with a@n.1


          1 Robert W. Fun