Grace Theological Journal 3.1 (1982) 81-88.

[Copyright 1982 Grace Theological Seminary; cited with permission;

digitally prepared for use at Gordon College]








Less frequent than other types of conditional sentences, second

class conditions are also more specialized in their meaning and more

restricted in their grammatical format. In these alone the verb tenses

used provide the formal key to their identification. The major exegeti-

cal question, and the only serious divergence on the part of gram-

marians, centers around these tenses.. This study concludes that the

tenses used were determined by normal aspectual considerations, not

by arbitrary rule of grammar.

* * *


SECOND class conditional sentences occur less frequently than

other types in the NT; there are only 47 examples.1 Called by some

"Contrary to Fact" or "Unreal",2 by others "Determined as Unful-

filled,") they enjoy more agreement on the part of the grammarians

than the other types and are less problem for the exegete.


1 As compared with more than 300 first class and about 250 third class. There are

no complete fourth class conditions in the NT. A listing of these 47 examples may be

had by combining the lists given in notes 16-19, plus the two exceptions listed in the

text below.

2 So commonly in the grammars of classical Greek: W. W. Goodwin, Greek

Grammar, rev. by. C. B. Gulick (Boston: Ginn, 1930) 296, Hadley and Allen, Greek

Grammar (New York: D. Appleton, 1890) 283, Adolph Kaegi, A Short Grammar

Classical Greek (St. Louis: B. Herder, 1914) 143, and H. W. Smyth, A Greek Grammar

(New York: American Book Co., 1916) 342. Among NT Greek grammars also: F. Blass

and A. DeBrunner, A Greek Grammar of the New Testament and Other Early Christian

Literature, trans. and rev. by Robert Funk (Chicago: University of Chicago, 1961) 182,

H. Dana and J. Mantey, A Manual Grammar of the Greek New Testament (New York:

Macmillan) 287, W. S. LaSor, Handbook of New Testament Greek (Grand Rapids:

Eerdmans, 1973) B223, H. P. V. Nunn, A Short Syntax of New Testament Greek (Cam-

bridge: Cambridge University, 1951) 117, and Nigel Turner, Syntax, Vol. 3 of A Grammar

of New Testament Greek (Edinburgh: T. & T. Clark, 1963) 91.

3 J. H. Moulton, An Introduction to the Study of New Testament Greek (New

York: Macmillan, 1955) 211, S. G. Green, A Handbook of the Grammar of the Greek

Testament (New York: Revell, n.d.) 283, A. T. Robertson, A Grammar of the Greek

New Testament in the Light of Historical Research (Nashville: Broadman, 1934) 1012,

W. D. Chamberlain, An Exegetical Grammar of the Greek New Testament (Grand

Rapids: Baker, 1941) 195.





Second class conditions are more formally structured than either

of the other types. Both first and third class show a characteristic

structure only in the protasis, but the second class shows a distinctive

pattern in both the protasis and apodosis; indeed, it is the apodosis

which clearly identifies it.

The protasis uses the conditional conjunction ei] with the verb in

the indicative mood. In this it is like the first class. But the second class

uses only past tenses,4 whereas the first class may use any tense. Thus,

theoretically, there can be ambiguity in the form of the protasis, but in

few cases does this cause confusion of identification.5

The apodosis of second class conditions also uses a past tense of

the indicative, usually6 with a@n in almost7 every instance, the apodosis

is a simple statement of a non-fact; what would be or would have been

but was not. This contrasts strongly with the great variety of apodosis

forms occurring in the first and third classes.

The negative in the protasis is almost always mh<, with only two

instances of ou]k.8 This gives many examples of ei] mh< coming together

where mh< is simply the negation of the clause. There are a few instances

where it seems to be ei] mh< = "except" or "unless.9 The negative of the

apodosis is always ou]k.10 Both mh< in the protasis and ou]k in the

apodosis are what we would expect. In the protasis, which states a

potential circumstance, that which might have been, mh< is used. Ou]k is


4 These are the secondary or augmented tenses of the indicative: the imperfect,

aorist, and pluperfect.

5 In about one-sixth of the first class conditions a past tense indicative verb is used

in the protasis, but the identification is unambiguous because the apodosis is not

compatible with the second class form. In a few instances (Acts 11:17, Rom 5:15, Eph

4:21, Rev 20: 15) the form of both the protasis and the apodosis could be second class,

but the sense is clearly not contrary to fact. Of course, this is not unnatural; a simple

condition (first class) can be used of the past as naturally as of the present and future


6 @An occurs in 36 examples; it is omitted in 11 instances. This tendency to omit a@n

is characteristic of koine Greek.

7 In one instance (Luke 19:42) the apodosis is not stated. In two instances (I Cor

12:17, 19) the apodosis is a rhetorical question implying the simple statement, "There

would be none."

8 Mh< occurs 11 times. The two occurrences of ou]k (Matt 26:24, Mark 14:21) are

actually parallel passages duplicating a single occurrence.

9 This phenomenon of ei] mh< = "except" or "unless" will be dealt with separately at

another time.

10 There is a negative apodosis in 23 of the 47 examples. Ou]k is used in 22 of them,

ou]d ] (ou] de< = "not even") in one (Heb 8:4).




natural in the apodosis, which expresses nothing doubtful or sub-

jective, but states matter-of-factly what actually would have been if the

condition had been true.



There seems to be no debate on the essential meaning of the

second class conditional sentence. It states a condition which as a

matter of fact has not been met and follows with a statement of what

would have been true if it had. An extended paraphrase in English

would be, "If this were the case, which it is not, then this would have

been true, which as a matter of fact, is not." The term "contrary to

fact" therefore is an accurate descriptive name for this type.11

It must be kept in mind in the use of this descriptive term that

"contrary to fact" has to do with the statement of the fact, not the

actual fact itself. The speaker states it as being contrary to fact; he may

or may not be correct in that statement. Of the 47 NT examples, 39 are

by Christ or by inspired writers of scripture; in every case, the

statement is also contrary to fact in actuality. In each of the other 8

examples, where the speakers were men liable to error, they spoke

what they believed to be contrary to fact; in two instances they were


A very significant comparison must be made here. In dealing with

the significance of the first class condition, this distinction between fact

and statement of fact sometimes has been used to explain those many

examples where the first class is used in obviously false or uncertain

statements.13 However, there is a drastic difference in this respect

between first and second class. In the first class examples where there is

a discrepancy between the actual fact and the statement of it, it is not a

matter of error or ignorance; it is almost always a deliberate statement

of what is known or considered by the speaker to be false. But in the

second class, there is not a single instance of stating something as

contrary to fact which is not so in the judgment of the speaker. He is

making what he considers a contrary-to-fact statement. There is no


11 A. T. Robertson's designation "Determined as Un-Fulfilled" seems also to be a

valid characterization. The problem with his system of classifying conditional sentences

lies in his designating the first class "Determined as Fulfilled," which understandably

has been misinterpreted as the opposite of the second class, therefore "True to Fact."

See my preceding article: "First Class Conditions: What Do They Mean?", GTJ 2

(1981) 79-80.

12 Luke 7:39, John 18:30.

13 See the discussion in my preceding article, "First Class Conditions," 77-78.




such thing as "assuming for the sake of argument" that a statement is

contrary to fact. To put it in another way, the first class condition is

not the opposite of the second class. It is not "true to fact" in the sense

that the second is contrary to fact.



In dealing with the significance of the tenses used, two factors

require consideration: first, the fact that only past tenses of the

indicative are used, and second, the question of the time relation



Only Past Tenses

Contrary-to-fact conditional sentences are the only type which

has tense limitation. Why? And why these tenses? The answer will help

to explain and support the meaning assigned to this type of con-


All conditional sentences by their very nature involve statements

which may or may not be true. That is what "if" means. The

uncertainty involved may be due to ignorance, supposition, choice,

course of events (I call it providence), or simple futurity. If the time

involved is either present or future, there is always this element of

uncertainty from the viewpoint of the human speaker (both Greek

and English are human languages). Only in past time has the uncer-

tainty become certainty by actual occurrence, and even then it is not

certain to the speaker until and unless he knows about it. The second

class condition is one which expresses the "would be" results of a past

condition known (or thought) to be unfulfilled or contrary to fact.

Very naturally, then, it uses only past tenses.

It is instructive to note that this usage is but one example of

what grammarians have called the "potential" or "unreal" indicative.

This idiom includes, beside the unreal conditional sentence, such

other uses of the augmented tenses of the indicative, with or without

a@n, as in courteous or polite language (Acts 25:22, Gal 4:20), in

expressions of necessity, obligation, possibility, and propriety (Luke

24:26, Acts 24: 19, 1 Cor 5:10), and in cautious statements and

impossible wishes (Rom 9:3). Even in English we use "ought,"

"would," "could"--past tense forms which are used in many of these

unreal statements.14


14 For a discussion of the idiom, consult the grammars: (classical) Goodwin and

Gulick, Greek Grammar, 283, 297, Kaegi, Short Grammar, 136, 137, Smyth, Greek

Grammar, 296; (NT) Dana and Mantey, Manual Grammar, 169, A. T. Robertson,

Grammar, 918-23, Turner, Syntax, 90-93.




Time Reference

Some grammarians have distinguished two time references in

second class conditions, indicated by the tense used in the protasis.15

It is claimed that the imperfect tense is used for a statement which is

presently contrary to fact, the aorist and pluperfect for a past

contrary-to-fact condition. Is this a valid distinction in NT Greek?

It should be noted that this, like all considerations dealing with

Greek tense, is more a matter of aspect or aktionsart than of time. By

the very nature of the case all contrary-to-fact conditions are to some

extent past in time. The decision that it is not fulfilled has already

been made before the sentence is uttered or written. "If you believed

Moses you would believe me" (John 5:46) is speaking of a present

situation which is not true; they are not at that moment believing.

The imperfect tense used is a durative tense. They are in a state of

unbelieving which is presently continuing but of course it has already

been in existence long enough to be known as untrue. If the aorist

had been used in this protasis the sense might have been, "If you had

(sometime in the past) exercised faith, you would have (now) believed


Most NT examples fit well into this distinction. All of those

using the aorist16 and the pluperfect17 are past in time reference,

properly expressed in English with a past perfect: "If it had been. ..

it would have been. . . ." The case is not quite so clear-cut with the

imperfect, but even here two-thirds of the examples fit the pattern,18

indicating a present time reference, "if it were. . . , it would be. . . ."

Of the nine apparent exceptions, seven19 are instances of the imperfect

of the verb ei]mi<. Since this verb has only one past tense (apparently


15 Dana and Mantey [289] make the strange assertion that "a contrary to fact

condition dealing with present time has the imperfect tense in both protasis and

apodosis . . . a contrary to fact condition dealing with past time has the aorist or

pluperfect tense in both protasis and apodosis," even though two of the examples they

cite show a mixed use, with different tenses in the two clauses. In view of the fact that

16 of the NT examples actually show such mixed tenses (9 examples have the imperfect

in the protasis with aorist or pluperfect in the apodosis; 7 have the reverse situation; all

but one seem to be past in time reference) this statement obviously is an overstatement.

If there is any relation between tense and time reference, it is the tense of the protasis

which must be the determining one.

16 There are 16 examples: Matt 11:21, 11:23, 12:7, 24:22, 26:24, Mark 13:20, 14:21,

Luke 10:13, 19:42, John 4:10, 15:20, 15:24, Rom 9:29, I Cor 2:8, Gal 3:21, Heb 4:8.

17 There are 4 examples: Matt 24:43, Luke 12:39, John 8: 19, Acts 26:32. John 19:11

is questionable. Cf. my treatment of this verse below.

18 15 out of 24 examples: Luke 7:39, John 5:46, 8:42, 9:33, 9:41, 15:19, 18:36, 19:11

(?), Acts 18:14, I Cor 11:31, 12:17, 12:19, Gal. 1:10, Heb 8:4, 8:7.

19 Matt 23:30, John 11:21, 11:32, 18:30, Gal. 4:15, I John 2:19. Also, in John 1.4:2

the verb is unexpressed but most naturally it would be h#n, the imperfect of ei]mi<.




the intrinsically durative aspect of this verb rendered unnecessary the

development of an aorist and pluperfect conjugation) it is conceivable

that grammatical constructions which normally called for those tenses

may have been met by substituting the imperfect. However, aside

from this rationalization, the basic aspect of the imperfect tense fits

perfectly in each of the seven cases. While the sense demands that the

time reference is past, the kind of action is durative in that past time.

The remaining two apparent exceptions to the general rule under

consideration may be explained in a similar way. In John 14:28, "if

you loved me, you would have rejoiced," it seems clear that the time

reference is past. Earlier in the verse Christ reminded them of his

impending departure and return and follows that statement with this

condition. He was clearly thinking of love as a durative state of being,

"if you were (at that time) loving me," rather than a specific act of

love. His use of the imperfect emphasizes this.

In Rom 7:7 the case is not quite so clear. First, it may be seen as

a present contrary-to-fact condition: "I would not (now) know lust if

the law were not continually saying. . . ." This would probably be

easiest grammatically. Even the verb in the apodosis is in sense an

imperfect, since the verb oi#da is a perfect form with a present

meaning and its pluperfect form is the corresponding imperfect. But

the sense resulting is impossible. Or, second, it may be seen as a past

contrary-to-fact condition: "I would not have known lust if the law

had not said. . ." If this is the sense, then the imperfect verb would

be calling attention to the durative aspect: "If the law were not

continually telling me. . . ," emphasizing the persistent influence of

Paul's exposure to law-teaching.

In summary, it seems generally to be true that an imperfect verb

in the protasis of a second class condition indicates a present-time

condition and an aorist or pluperfect verb indicates a past-time

condition. The few apparent exceptions are examples where the

durative nature of the past-time condition is emphasized by the use of

the imperfect. But the existence of a considerable number of excep-

tions points rather to the conclusion that this "rule" works because of

the durative sense of the imperfect rather than because it was a

required structural pattern. It is better to approach the meaning by

giving' attention to the aspect of the tenses used rather than to an

imagined rule.


Other Noteworthy Examples

Individual consideration needs to be given to a few examples

which show some unusual characteristics.

Luke 17:6. "If you have faith. . . you would be saying. . ." The

protasis has ei] with a present indicative verb and is therefore a first



class condition. But the apodosis has a@n with an imperfect verb,

which fits the second class pattern. Thus it is cited as an example of

what grammarians sometimes call a "mixed condition.20 There is

nothing inherently unlikely about such a situation, and Nigel Turner

well explains its peculiar appropriateness in this instance21 as express-

ing a subtle politeness which avoided the harshness of saying, "If you

had faith (which you do not) . . . ," the blunt meaning which would

have resulted if he had used the full second class form.22 However, it

is possible to see an entirely different solution to this unusual

construction. It is clear that the protasis is first class, a simple

condition implying nothing as to whether Jesus' hearers actually had

faith, and thus neither congratulating them nor criticizing them.

Furthermore, it is clear from multitudes of examples that the apodosis

of a first class condition may be of any form (declarative, hortatory,

command, promise, rhetorical question, wish, etc.). A normal usage

of a@n with the imperfect which is not a second class apodosis does

exist; it may well be the "potential" use of past tense indicatives for

courteous or polite language or to express present necessity, obliga-

tion, possibility, or propriety.23 Applying this grammatical usage to

this passage, the sense becomes, in expanded paraphrase, "If you

have faith, you could say to this mountain. . . ," or, "it would be right

and proper for you to say. . . ,"or, "if you have faith there is nothing

you cannot ask for."

John 8:39. "If you are Abraham's children, you would be doing

the works of your father" may also be an example of a mixed

condition, with a first class protasis to soften the harshness of the

statement. The textual tradition would suggest this understanding,

whether the United Bible Society preferred reading e]poiei?te or the

Byzantine text a}n e]poiei?te is followed. In this instance, the explana-

tion of the apodosis as a potential indicative, suggested for the

preceding example, is not agreeable to the sense. Another reading, the

imperative poiei?te, followed by the NASB, would be a regular first

class condition.

Heb 11:15. "If they were remembering the place from which they

went out, they would have an opportunity to return" also involves a

textual variation. The apodosis is clearly of the second class. In the


20 A. T. Robertson, Grammar, 1022.

21 N. Turner, Grammatical Insights into the New Testament (Edinburgh: T. &. T.

Clark, 1965) 51-52.

22 See my note on Turner's questionable understanding of the significance of the

first class condition as reflected in his treatment of this passage in my preceding article,

"First Class Conditions," 81, n. 17.

23 See my discussion of this idiom earlier in this article. Also, R. Law, "Imperfect

of 'Obligation' etc., in the N.T.," ExpT 30 (1919), 330ff.





UBS text the protasis has its verb in the present indicative and is thus

of the first class. But the Byzantine text, accepted here by Westcott

and Hort, has the imperfect tense, making the whole a normal second

class condition. Here the time reference is actually past, even though

imperfects, according to the rule discussed earlier, would be con-

sidered by some to signal a present contrary to fact. Perhaps the

writer uses this "present" form from the same vantage point as in the

preceding verse, which uses the "historical present" to express vividly

a past situation. Or perhaps the present time reference in both verses

is the "gnomic present"; it is always or characteristically true that if

someone keeps looking back there are opportunities to go back. The

use of the durative imperfect stresses the continuing situation: "if they

were remembering. . . they would be having continuing opportunity

to return."

John 19:11. "You would have no authority over me if it had not

been given you from above." The problem here also is the time

reference. If the verb of the protasis is taken as h#n dedome<non, a

periphrastic pluperfect, then the time reference would be past, "If it

had not been given. . ." If the verb is understood to be h#n alone, with

the perfect participle functioning as a predicate adjective, then the

imperfect verb might be signaling a present contrary to fact: "if it

were not (now) an authority which has been given you. . ." It is

probably a distinction without a difference. In either case, the imper-

fect in the apodosis indicates the present situation.




This material is cited with gracious permission from:

Grace Theological Seminary

200 Seminary Dr.

Winona Lake, IN 46590

Please report any errors to Ted Hildebrandt at: