Albert Barnes Commentary Hebrews 9:16

"For where a testament is, there must of necessity be the death of him that made it." — Hebrews 9:16 (ASV)

For where a testament is. This is the same word, diayhkh, which in Hebrews 8:6 is rendered covenant. For the general meaning of the word, see the commentary of Barnes on Hebrews 8:6.

So much depends, however, on the meaning of the word—not only in the interpretation of this passage but also of other parts of the Bible—that it may be proper to explain it here more at length. The word diayhkh—occurs in the New Testament thirty-three times. It is translated covenant in the common version in Luke 1:72; Acts 3:26; Acts 7:8; Romans 9:4; Romans 11:27; Galatians 3:15, 17; Galatians 4:24; Ephesians 2:12; Hebrews 8:6, 8, 9, 10; Hebrews 9:4; Hebrews 10:16; Hebrews 12:24; and Hebrews 13:20.

In the remaining places it is rendered testament: Matthew 26:28; Mark 14:24; Luke 22:20; 1 Corinthians 11:25; 2 Corinthians 3:6, 14; Hebrews 7:22; Hebrews 9:15–17, 20; and Revelation 11:19.

In four of those instances (Matthew 26:28; Mark 14:24; Luke 22:20; and 1 Corinthians 11:25), it is used with reference to the institution or celebration of the Lord’s Supper.

In the Septuagint, it occurs nearly three hundred times. In considerably more than two hundred of these instances, it is the translation of the Hebrew word HEBREW Berith. In one instance, Zechariah 11:14, it is the translation of the word brotherhood; once, Deuteronomy 9:5, of HEBREW—word; once, Jeremiah 34:18, of “words of the covenant;” once, Leviticus 26:11, of tabernacle; once, Exodus 31:7, of testimony. It occurs once, Ezekiel 16:8, where the reading of the Greek and Hebrew text is doubtful; and it occurs three times, 1 Samuel 11:2; 1 Samuel 20:8; 1 Kings 8:9, where there is no corresponding word in the Hebrew text.

From this use of the word by the authors of the Septuagint, it is evident that they regarded it as the proper translation of the Hebrew HEBREW Berith, and as conveying the same sense that word does. It cannot be reasonably doubted that the writers of the New Testament were led to use the word, in part at least, by the fact that they found it occurring so frequently in the version in common use. However, it also cannot be doubted that they regarded it as fairly conveying the sense of the word HEBREW Berith. On no principle can it be supposed that inspired and honest men would use a word, when referring to transactions in the Old Testament, which did not fairly convey the idea that the writers of the Old Testament meant to express.

The use being thus regarded as settled, there are some facts in reference to it which are of great importance in interpreting the New Testament and in understanding the nature of the “covenant” which God makes with man. These facts are the following:

1. The word diayhkh diatheke—is not that which properly denotes compact, agreement, or covenant. That word is sunyhkh—syntheke—or, in other forms, sunyesiv and sunyesia. Or, if the word diatheke is used in that meaning, it is only remotely and as a secondary meaning. See Passow; compare the Septuagint in Isaiah 28:15; Isaiah 30:1; Daniel 11:6; Wisdom 1:16; 1 Maccabees 10:26; 2 Maccabees 13:25; 2 Maccabees 14:26. It is not the word which a Greek would have employed to denote a compact or covenant; he would have employed it to denote a disposition, ordering, or arrangement of things, whether of religious rites, civil customs, or property. Or, if used with reference to a compact, it would have been with the idea of an arrangement or ordering of matters, not with the primary notion of an agreement with another.

2. The word properly expressive of a covenant or compact, sunyhkh, is never used in the New Testament. In all the allusions to the transactions between God and man, this word never occurs. From some cause, the writers and speakers in the New Testament seem to have supposed that the word would leave an impression which they did not wish to leave. Though it might have been supposed that, in speaking of the various transactions between God and man, they would have selected this word, yet with entire uniformity they have avoided it. Not one of them—though the word diayhkh diatheke—has been used by no less than six of them—has been betrayed in a single instance into the use of the word sunyhkh syntheke, nor has differed from the other writers in the language employed. This cannot be supposed to be the result of concert or collusion but must have been founded on some reason which operated equally on all their minds.

3. In like manner, and with similar remarkable uniformity, the word sunyhkh syntheke—is never used in the Septuagint with reference to any arrangement or “covenant” between God and man. Once indeed in the Apocrypha, and only once, it is used in that sense. In the three only other instances in which it occurs in the Septuagint, it is with reference to compacts between man and man (Isaiah 28:16; Isaiah 30:1; Daniel 11:6).

   This remarkable fact, that the authors of that version never use the word to denote any transaction between God and man, shows that there must have been some reason for it which acted on their minds with entire uniformity.

4. It is no less remarkable that neither in the Septuagint nor the New Testament is the word diayhkh—diatheke—ever used in the sense of will or testament, unless it be in the case before us. This is conceded on all hands and is expressly admitted by Professor Stuart (Commentary on Hebrews, p. 439), though he defends this use of the word in this passage.

A very important inquiry presents itself here which has never received a solution generally regarded as satisfactory. It is why the word diayhkh—diatheke—was selected by the writers of the New Testament to express the nature of the transaction between God and man in the plan of salvation. It might be said, indeed, that they found this word uniformly used in the Septuagint, and that they employed it as expressing the idea which they wished to convey with sufficient accuracy. But this is only removing the difficulty one step further back. Why did the Seventy adopt this word? Why did they not rather use the common and appropriate Greek word to express the notion of a covenant?

A suggestion on this subject has already been made in the notes on Hebrews 8:6. (Compare Biblical Repository, vol. 20, p. 55.) Another reason may, however, be suggested for this remarkable fact which is liable to no objection. It is that, in the apprehension of the authors of the Septuagint and of the writers of the New Testament, the word diayhkh diatheke—in its original and proper meaning, fairly conveyed the sense of the Hebrew word HEBREW Berith, and that the word sunyhkh syntheke—or compact, agreement, would not express that. And that they never meant to be understood as conveying the idea either that God entered into a COMPACT or COVENANT with man, or that He made a WILL. They meant to represent Him as making an arrangement, a disposition, an ordering of things, by which His service might be kept up among His people, and by which men might be saved; but they were equally remote from representing Him as making a compact, or a will.

In support of this, the following may be alleged:

1. The remarkable uniformity with which the word diayhkh diatheke—is used, showing that there was some settled principle from which they never departed; and

2. Secondly, the meaning of the word itself. Professor Stuart has, undoubtedly, given the accurate original sense of the word. “The real, genuine, and original meaning of diayhkh diatheke—is, arrangement, disposition, or disposal of a thing” (p. 440). The word from which it is derived—diatiyhmi—means to place apart or asunder, and then to set, arrange, or dispose in a certain order (Passow).

From this original meaning is derived the use which the word has, with singular uniformity, in the Scriptures. It denotes the arrangement, disposition, or ordering of things which God made in relation to mankind, by which He designed to keep up His worship on earth and to save the soul.

It means neither covenant nor will; neither compact nor legacy; neither agreement nor testament. It is an arrangement of an entirely different order from either of them, and the sacred writers, with a uniformity which could have been secured only by the presiding influence of the One Eternal Spirit, have avoided the suggestion that God made with man either a compact or a will.

We have no word which precisely expresses this idea; and hence our conceptions are constantly floating between a compact and a will, and the views we have are as unsettled as they are unscriptural. The simple idea is that God has made an arrangement by which His worship may be celebrated and souls saved.

Under the Jewish economy, this arrangement assumed one form; under the Christian, another. In neither was it a compact or covenant between two parties in such a sense that one party would be at liberty to reject the terms proposed; in neither was it a testament or will, as if God had left a legacy to man. But in both, there were some things in regard to the arrangement such as are found in a covenant or compact.

One of those things—equally appropriate to a compact between man and man, and to this arrangement the apostle refers to here—is that it implied in all cases the death of the victim. If these remarks are well founded, they should be allowed materially to shape our views in the interpretation of the Bible.

Whole treatises of divinity have been written on a mistaken view of the meaning of this word—understood as meaning covenant. Volumes of angry controversy have been published on the nature of the “covenant” with Adam, and on its influence on his posterity. The only literal covenant which can be supposed in the plan of redemption is that between the Father and the Son—though even the existence of such a covenant is rather the result of devout and learned imagining than of any distinct statement in the volume of inspiration.

The simple statement there is that God has made an arrangement for salvation, the execution of which He has entrusted to His Son, and has proposed it to man to be accepted as the only arrangement by which man can be saved, and which he is not at liberty to disregard.

There has been much difference of opinion in reference to the meaning of the passage here, and to the design of the illustration introduced. If the word used—diayhkh—means testament, in the sense of a will, then the sense of that passage is that “a will is of force only when he who made it dies, for it relates to a disposition of his property after his death.” The force of the apostle’s remark then would be that the fact that the Lord Jesus made or expressed His will to mankind implied that He would die to confirm it; or that since, in the ordinary mode of making a will, it was of force only when he who made it was dead, therefore it was necessary that the Redeemer should die in order to confirm and ratify that which He made. But the objections to this, which appears to have been the view of our translators, seem to me to be insuperable. They are these:

1. The word diayhkh—diatheke—is not used in this sense in the New Testament elsewhere. See the remarks above.

2. The Lord Jesus made no such will. He had no property, and the commandments and instructions which He gave to His disciples were not of the nature of a will or testament.

3. Such an illustration would not be pertinent to the apostle’s design, or in keeping with his argument. He is comparing the Jewish and Christian dispensations, and the point of comparison in this chapter relates to the question about the efficacy of sacrifice in the two arrangements. He showed that the arrangement for bloodshedding by sacrifice entered into both; that the high priest of both offered blood as an expiation; that the holy place was entered with blood; and that consequently there was death in both the arrangements or dispensations.

   The former arrangement or dispensation was ratified with blood, and it was equally proper that the new arrangement should be also. The point of comparison is not that Moses made a will or testament which could be of force only when he died, and that the same thing was required in the new dispensation; but it is that the former covenant was ratified by blood, or by the death of a victim, and that it might be expected that the new dispensation would be confirmed—and that it was, in fact, confirmed in the same manner. In this view of the argument, what pertinence would there be in introducing an illustration respecting a will and the manner in which it became efficient? See the notes on Hebrews 9:18.

It seems clear, therefore, to me, that the word rendered testament here is to be taken in the sense in which it is ordinarily used in the New Testament. The opinion that the word here means such a Divine arrangement as is commonly denoted a “covenant,” and not testament, is sanctioned by not a few names of eminence in criticism, such as Pierce, Doddridge, Michaelis, Steudel, and the late Dr. J.P. Wilson. Bloomfield says that the connection here demands this. The principal objections to this view are:

1. That it is not proved that no covenants or compacts were valid, except such as were made by the intervention of sacrifices.

2. That the word rendered testator, diayemenov—cannot refer to the death of an animal slain for the purpose of ratifying a covenant, but must mean either a testator or a contractor, i.e., one of two contracting parties.

3. That the word rendered dead (Hebrews 9:17)—nekroiv—means only dead men, and never is applied to the dead bodies of animals. See Stuart on Hebrews, p. 442.

These objections to the supposition that the passage refers to a covenant or compact, Professor Stuart says are, in his view, insuperable, and they are certainly entitled to grave consideration. Whether the view above presented is one which can be sustained, we may be better able to determine after an examination of the words and phrases which the apostle uses. Those objections which depend wholly on the philological argument derived from the words used will be considered, of course, in such an examination. It is to be remembered at the outset:

1. That the word diayhkh—diatheke—is never used in the New Testament in the sense of testament or will, unless in this place;

2. That it is never used in this sense in the Septuagint; and

3. That the Hebrew word HEBREW—Berith—never has this meaning. This is admitted. See Stuart on Hebrews, pp. 439, 440.

It must require very strong reasons to prove that it has this meaning here, and that Paul has employed the word in a sense differing from its uniform meaning elsewhere in the Bible. Compare, however, the remarks of Professor Stuart in Biblical Repository, vol. 20, p. 364.

There must also of necessity be. Anagkh—That is, it is necessary in order to confirm the covenant, or it would not be binding in cases where this did not occur. The necessity in the case is simply to make it valid or obligatory. So we say now, there must “necessarily” be a seal, or a deed would not be valid. The fair interpretation of this is that this was the common and established custom in making a “covenant” with God, or confirming the arrangement with Him in regard to salvation. To this it is objected (see the first objection above) that “it is yet to be made out that no covenants were valid except those by the intervention of sacrifices.” In reply to this, we may observe:

1. That the point to be made out is not that this was a custom in compacts between man and man, but between man and his Maker.

   There is no evidence, as it seems to me, that the apostle alludes to a compact between man and man. The mistake on this subject has arisen partly from the use of the word “testament” by our translators, in the sense of will—supposing that it must refer to some transaction relating to man only; and partly from the insertion of the word “men” in Hebrews 9:17, in the translation of the phrase—epi nekroiv, “upon the dead,” or “over the dead.” But it is not necessary to suppose that there is a reference here to any transaction between man and man at all, as the whole force of the illustration introduced by the apostle will be retained if we suppose him speaking only of a covenant between man and God. Then his assertion will be simply that, in the arrangement between God and man, there was a necessity of the death of something, or of the shedding of blood in order to ratify it. This view will save the necessity of proof that the custom of ratifying compacts between man and man by sacrifice prevailed. Whether that can be made out or not, the apostle’s assertion may be true, that in the arrangement which God makes with man, sacrifice was necessary in order to confirm or ratify it.

2. The point to be made out is not that such a custom is or was universal among all nations, but that it was the known and regular opinion among the Hebrews that a sacrifice was necessary in a “covenant” with God. In the same way, if we should say that a deed was not valid without a seal, it would not be necessary to show this in regard to all nations, but only that it is the law or the custom in the nation where the writer lived, and at the time when he lived.

   Other nations may have very different modes of confirming or ratifying a deed, and the same nation may have different methods at various times. The fact or custom to which I suppose there is allusion here is that of sacrificing an animal to ratify the arrangement between man and his Maker, commonly called a “covenant.” In regard to the existence of such a custom, particularly among the Hebrews, we may make the following observations. It was the common mode of ratifying the “covenant” between God and man. That was done over a sacrifice, or by the shedding of blood. So the covenant with Abraham was ratified by slaying a heifer, a she-goat, a ram, a turtle-dove, and a young pigeon.

   The animals were divided and a burning lamp passed between them (Genesis 15:9, 18). So the covenant made with the Hebrews in the wilderness was ratified in the same manner (Exodus 24:6 and following). Thus, in Jeremiah 34:18, God speaks of the “men that had transgressed his covenant which they had made before him when they cut the calf in twain and passed between the parts thereof.” See also Zechariah 9:11.

   Indeed, all the Jewish sacrifices were regarded as a ratification of the covenant. It was never supposed that it was ratified or confirmed in a proper manner without such a sacrifice. Instances occur, indeed, in which there was no sacrifice offered when a covenant was made between man and man (Genesis 24:9; Deuteronomy 25:7, 9; Ruth 4:7); but these cases do not establish the point that the custom did not prevail of ratifying a covenant with God by the blood of sacrifice. Further, the terms used in Hebrew in regard to making a covenant with God prove that it was understood to be ratified by sacrifice, or that the death of a victim was necessary.

HEBREW. Berith, “to cut a covenant”—the word HEBREW karath meaning to cut, to cut off, to cut down; the allusion being to the victims offered in sacrifice and cut in pieces on occasion of entering into a covenant. See Genesis 15:10; Jeremiah 34:18, 19.

The same idea is expressed in the Greek phrases orkia temnein, temnein spondav, and in the Latin icere faedus. Compare Virgil, Aeneid 8.641:

Et caesa jungebant foedera porca.

These considerations show that it was the common sentiment, alike among the Hebrews and the heathen, that a covenant with God was to be ratified or sanctioned by sacrifice. The statement of Paul here is that the death of a sacrificial victim was needful to confirm or ratify such a covenant with God.

It was not secure, or confirmed, until blood was thus shed. This was well understood among the Hebrews: that all their covenant transactions with God were to be ratified by a sacrifice. Paul says that the same principle must apply to any arrangement between God and men. Hence he goes on to show that it was necessary that a sacrificial victim should die in the new Covenant which God established with man through the Mediator.

See Hebrews 9:23. This I understand to be the sum of the argument here. It is not that every contract made between man and man was to be ratified or confirmed by a sacrifice—for the apostle is not discussing that point. Rather, it is that every similar transaction with God must be based on such a sacrifice, and that no covenant with Him could be complete without such a sacrifice.

This was provided for in the ancient dispensation by the sacrifices which were constantly offered in their worship; in the new, by the one great Sacrifice offered on the cross. Hence all our approaches to God are based on the supposition of such a sacrifice and are, as it were, ratified over it. We ratify or confirm such a covenant arrangement, not by offering the sacrifice anew, but by recalling it in a proper manner when we celebrate the death of Christ, and when, in view of His cross, we solemnly pledge ourselves to be the Lord’s.

The death of the testator. According to our common version, the death of him who makes a will. But if the views above expressed are correct, this should be rendered “the covenanter,” or “the victim set apart to be slain.” The Greek will admit of the translation of the word diayemenov—diathemenos—by the word covenanter, if the word diayhkh diatheke—is rendered covenant. To such a translation here as would make the word refer to a victim slain in order to ratify a covenant, it is objected that “the word has no such meaning anywhere else. It must either mean a testator, or a contractor, i.e., one of two covenanting parties. But where is the death of a person covenanting made necessary in order to confirm the covenant?” (Professor Stuart, in loc.) To this objection I remark respectfully:

1. That the word is never used in the sense of testator, either in the New Testament or the Old, unless it be here. It is admitted of the word diayhkh, diatheke—by Professor Stuart himself—that it never means will, or testament, unless it be here; and it is equally true of the word used here that it never means one who makes a will. If, therefore, it should be that a meaning quite uncommon, or wholly unknown in the usage of the Scriptures, is to be assigned to the use of the word here, why should it be assumed that that unusual meaning should be that of making a will, and not that of confirming a covenant?

2. If the apostle used the word diayhkh—diatheke—in the sense of a covenant in this passage, nothing is more natural than that he should use the corresponding word diayemenov—diathemenos—in the sense of that by which a covenant was ratified. He wished to express the idea that the covenant was always ratified by the death of a victim—a sacrifice of an animal under the law, and the sacrifice of the Redeemer under the gospel—and no word would so naturally convey that idea as the one from which the word covenant was derived.

   It is to be remembered, also, that there was no word to express that thought. Neither Hebrew nor Greek furnished such a word; nor do we now have any word to express that thought, but are obliged to use circumlocution to convey the idea. The word covenanter would not do it; nor the words victim or sacrifice.

   We can express the idea only by some phrase like this—“the victim set apart to be slain to ratify the covenant.” But it was not an unusual thing for the apostle Paul to make use of a word in a sense quite peculiar to himself. Compare 2 Corinthians 4:17.

3. The word diatiyhmi diatithemi—properly means, to place apart, to set in order, to arrange. It is rendered appoint in Luke 22:29; made and make, with reference to a covenant, in Acts 3:25; Hebrews 8:10; Hebrews 10:16.

   It occurs nowhere else in the New Testament, except in the passage before us. The idea of placing, laying, disposing, arranging, etc., enters into the word—as to place wares or merchandise for sale, to arrange a contract, etc. See Passow. The fair meaning of the word here may be whatever goes to arrange, dispose, or settle the covenant, or to make the covenant secure and firm.

   If the reference is to a compact, it cannot relate to one of the contracting parties, because the death of neither is necessary to confirm it. But it may refer to that which was well known as an established opinion: that a covenant with God was ratified only by a sacrifice.

   Still, it must be admitted that this use of the word is not elsewhere found. The only material question is whether it is to be presumed that the apostle would employ a word in a single instance, in a peculiar meaning, where the connection would not render it difficult to be understood.

   This must be admitted, that he might, whichever view is taken of the meaning of this passage; for, on the supposition that he refers here to a will, it is conceded that he uses the word in a sense which does not once occur elsewhere either in the Old Testament or the New.

   It seems to me, therefore, that the word here may, without impropriety, be regarded as referring to the victim that was slain in order to ratify a covenant with God; and that the meaning is that such a covenant was not regarded as confirmed until the victim was slain. It may be added that the authority of Michaelis, Macknight, Doddridge, Bloomfield, and Dr. J.P. Wilson is a proof that such an interpretation cannot be a very serious departure from the proper use of a Greek word.