Hebrews 2:3-4 is one of the “usual suspects” that is rounded up from time to time, the allegation being that it supports Cessationism. Does the evidence really indicate this? No, not really, but Biblical material that may be used to back this point of view is so scarce that I suppose you have to take what you can get.
If facts are against you, argue the law; if the law is against you, argue the facts; if both the law and the facts are against you, you need Daniel Baird Wallace, PhD. The eminent Greek scholar has a short article on the topic entitled “Hebrews 2:3-4 and the Sign Gifts.” I think I do not overstate when I say that the article is technically dense and not all that easy to follow. Think of Peter on Paul: “There are some things in them that are hard to understand” (2 Pet. 3:16). Its very opacity, however, serves a rhetorical purpose: it discourages critical analysis of his arguments and encourages an ipse dixit acceptance based on Dr. Wallace’s well-deserved reputation. This explains citations I’ve seen of the article that take an “Emperor’s New Clothes” approach; “okay, I don’t get it, but Dr. Wallace is a smart guy, and he says it’s so.”
Yet, it is important to recognize that Dr. Wallace never claims to find Cessationism actually taught from the passage, but merely that it “seems to involve some solid inferences that the sign gifts had for the most part ceased.” What Dr. Wallace is offering here does not seem to be so much his own expert opinion, but something akin to a legal brief: selected arguments that could be made from the evidence, without necessarily personally endorsing them as valid. How “solid” then are these inferences? Not very, I’m afraid, as this post intends to demonstrate. To say that the case is not very strong is no knock on Dr. Wallace. Who better to make the best possible case that can be made from the text? Yet even he cannot make bricks without clay.
So I find myself once again taking issue with Dr. Wallace, whom as I have repeatedly stated, I hold in the highest esteem. However, the conclusion of his article states:
I do not pretend to think that this sole text solves the problem of the duration of the sign gifts. But whatever one’s views of such gifts, this passage needs to be wrestled with.
Taking this as an invitation to respond meaningfully, I then accept graciously and offer my analysis in the spirit of fraternal dialogue.
Let us begin by citing the passage in its context, starting with verse 1 of the chapter, with the portion of the passage that Dr. Wallace treats underlined:
Therefore we must pay much closer attention to what we have heard, lest we drift away from it. For since the message declared by angels proved to be reliable, and every transgression or disobedience received a just retribution, how shall we escape if we neglect such a great salvation? It was declared at first by the Lord, and it was attested to us by those who heard, while God also bore witness by signs and wonders and various miracles and by gifts of the Holy Spirit distributed according to his will.
1. My first objection is with the phrase “sign gifts” in Dr. Wallace’s title. This has the effect of injecting a concept foreign to the text itself. The passage refers to “gifts of the Holy Spirit” as a class, and not to any subset of these. Dr. Wallace never presents an argument that the term here (merismois, not the more familiar charismata) is meant to indicate only “certain spiritual gifts,” but merely dives in using that phase. More than a little gratuitous, this is “assuming facts not in evidence.” In fact, since the author explicitly mentions signs, wonders, and miracles, and then gifts of the Spirit, it would seem reasonable to expect these last at least to include “gifts” that are not also “signs.” At any rate, it is hard to see how any cogent Cessationism argument from this passage would not take down teaching and mercy along with prophecy and healing.
2. His argument begins by countering a Continuationist argument in a way that is hard not to characterize as “spin.” Cessationists assert that God empowered certain activities through His Church early on and then deliberately ceased to do so at a certain point of time. Is it illegitimate to ask “where is that in the Bible?” His response (which takes some force from the loaded term “prooftext”) is that we would hardly expect such a statement, given that the NT writers (a) themselves exercised the gifts and (b) anticipated the Parousia in their lifetime. I don’t dispute Dr. Wallace’s logic but isn’t this then essentially an admission that the cessation of particular spiritual gifts is not something that New Testament actually teaches.
3. The meat of Dr. Wallace’s case, however, involves syntactic consideration of the grammatical unit in which the phrase “gifts of the Holy Spirit” is found. This is something called a “genitive absolute,” and the ESV renders it: “while God also bore witness by signs and wonders and various miracles and by gifts of the Holy Spirit distributed according to his will.” Dr. Wallace takes pains to argue that its participle sunepimarturountos, “co-testifying,” bears a temporal adverbial relationship to the immediately preceding finite verb ebebaiothe, “was confirmed.” This latter is in the aorist tense and has unquestionable reference to past time. It is true that the participle generally piggy-backs on the tense of the main verb.
Dr. Wallace ruminates at length on the proper place to connect the genitive absolute semantically. He does this rather needlessly, though, I think. He is certainly correct in his conclusion here, and there are several other reasons for it that he doesn’t mention. At one point though he does suggest an unlikely alternative, with a notation that it would align with Continuationism. I don’t know whether any Continuationist ever did argue for that analysis, but it lets him appear to shoot down a Cessationist argument, albeit a phantom one.
There are a number of significant points in this argument which require comment, but one almost makes all others moot. The most important objection is made by Dr. Wallace himself, albeit relegated to a footnote at the bottom of the second page:
The aorist indicative means “it happened,” but we cannot legitimately extrapolate from that a meaning, “and it doesn’t happen now.” The aorist can’t be used to state a negative in the present time.
Having assured us we cannot do it, he then proceeds to do it anyway. Oddly, what “we cannot legitimately extrapolate” in the footnote, becomes the chief of his “solid inferences.”
He offers examples of cases in which the aorist does in fact refer to only-past-not-future events (and at least the one about the finished work of Christ is valid). So we know that a past event can be once-for-all, but this is no indication that events behind the aorist in this passage are. In fact, Dr. Wallace’s second footnote specifically refutes the idea, since he states that John’s writing of Revelation was yet future.
4. A genitive absolute is a structure derived from the circumstantial participle. Dr. Wallace describes the circumstantial participle and specifically the genitive absolute as expressing “adverbial relations.” I understand why he makes this statement, but I don’t think this is the best way to describe the function of the circumstantial participle in general terms. The circumstantial participle serves as part of the syntactic scheme by which Greek puts together related predications within a sentence. Semantically, this grants to each predication in a sentence a value relative importance to the whole, something linguists call prominence. Generally, a sentence has a single finite verb, the central predication. Either before this finite verb, after, or both, there are associated participles that typically, though not exclusively, encode predications of secondary rank, that is with lower prominence.
Greek stongly prefers this hierarchical system of subordinating, known as hypotaxis, rather than coordinating, or syntaxis. By contrast Hebrew greatly prefers syntaxis. You can see this in the heavy use of “and” in Gospel of Mark, probably the most Hebrew-like Greek in the New Testament. Hebrews, on the other hand, is (paradoxically?) the least Hebrew-like, and the Greek in the New Testament that most nearly reaches Classical Greek style. It thus favors hypotaxis and sophisticated use of participles.
English preference falls somewhere between Hebrew and Greek. This means that sometimes what English would express by coordination, Greek expresses with a participle. The best known example I can think of is the “attendant circumstances” usage Matthew 28:19 where the participle/finite verb sequence is very correctly rendered by syntaxis in English: “Go therefore and make disciples.” Trying to imagine some adverbial function in the participle here, such as temporal, is simply misunderstanding the syntax.
Of course the relation of the circumstantial participle may be and often is one we would categorize as adverbial, such as time, means, manner, or reason (when, how, or why). These are often used where in English we use a subordinate clause, introduced by a subordinating conjunction. They are often translated this way. However, Greek uses subordinate clauses in addition to circumstantial participles. So what’s the difference? The chief distinction is this: a coordinating conjunction specifically indicates the semantic relation. A circumstantial participle is understood to be dependant, but the nature of the relation is not indicated. This is because it is either (a) self-evident or (b) multiple, or (c) just not important.
The specific relation is only discerned from context. In any case whatever the relation of the circumstantial participle may be determined to be, that relationship is not a high-ranking bit of information, else it would not be left implicit. What this means in this case is that even if the participle sunepimarturountos should best be understood as temporal, the text does not treat this as a very important fact. Accordingly, an analysis that does is distorting the text’s own force of argument.
5. The genitive absolute is a specific type of circumstantial participle, one that supplies (in the genitive case) a semantic “subject” for the participle. This allows that subject to be different from the subject of the main finite verb. Typically also that subject would appear nowhere else in the sentence, though this is not a firm rule. The genitive absolute has a range of usage from the common to the rather recherché. The latter is exemplified by this passage, Hebrews 2:1-4, which is as elevated an example of Greek as any in the New Testament.
This brings us to what is, without question, the most misleading assertion in Dr. Wallace’s article. He describes the function of the genitive absolute as “usually of a temporal nature,” and further specifies that “over 90% of genitive absolute constructions are temporal.” I have no reason to doubt that this is factual, but without resorting to Mark Twain’s quip about statistics, let me state in no uncertain terms that percentages is not the way to exegete the function of a word. It would be all too easy to jump from “usually” to “probably” in this passage. That would be a mistake. Allow me to explain.
I mentioned earlier the range of use for the genitive absolute. To understand what I mean, I would like to refer you to a table that lays out the distribution of genitive absolutes by New Testament books. It was compiled by Lois K. Fuller and appears in her article “The ‘Genitive Absolute’ in New Testament/Hellenistic Greek: A Proposal for Clearer Understanding.”
She lists a total of 312 genitive absolute constructions in the New Testament. Strikingly, these are disproportionately found in the narrative books. Expository genre, such as the epistles contain relatively few. Among these Hebrews contains the most. Narrative literature, the gospels and Acts alone contain 264 of the 312 usages, or some 85%. Note the similarity to Dr. Wallace’s figure of 90%. This is due to the way the genitive absolute is used in narrative literature.
This is where the “usually” temporal usage comes in. Time reference is of the essence of narrative structure. Events follow one another in what linguists call contingent temporal succession. The genitive absolute comes in handy here, useful in making transitions, indicating a setting in which one event crosses another event in time. This is its dirt-simple use, employed even by writers who make no claim to refinement of style, such as Mark. In the following examples, I underline where the genitive absolute is located.
Now after Jesus was born in Bethlehem of Judea in the days of Herod the king, magi from the east arrived in Jerusalem… (Matt. 2:1, NASB)
When evening came, after the sun had set, they began bringing to Him all who were ill and those who were demon-possessed. (Mark 1:32, NASB)
As they were speaking to the people, the priests and the captain of the temple guard and the Sadducees came up to them (Acts 4:1, NASB).
This relatively common, temporal, type of genitive absolute is easy to identify. It typically occurs in initial position (“usually,” Dr. Wallace states in Beyond the Basics, p. 655), and in context it clearly serves as a transition, introducing the setting into which another action occurs.
The structure of Heb. 2:4 could hardly be more unlike these. It is sentence final, not initial, and there is no hint of a transition. It does not constitute a setting for another action. There is no reason beyond citing statistics to take it as temporal.
6. Dr. Wallace himself has this to say about temporal adverbial participles in Beyond the Basics:
As we have said, the temporal participle answers the question When? As well, if a particular adverbial participle is to be labeled as temporal, this should be the primary element the author wishes to stress… Therefore, once you have identified the temporal force of the participle, you should then go on and ask whether another more specific semantic value is intended. (p. 624)
Does Dr. Wallace do this himself? Not that I can see. He simply relies on the statistical data of the genitive absolute in very different contexts. Does the “God also bearing witness…” unit actually serve to answer the question When? Does it delimit the time frame in which the gospel “was attested to us by those who heard?” I submit that this would be a difficult case to make, to see time as its primary function.
7. Can we in fact determine a more specific semantic function? Yes, but to do so we have to back away from the tree and take a look at the forest. Surprisingly, Dr. Wallace does not do much to consider the context, which is the single most important factor for determining and the semantic contribution of each individual part to the overall flow of argument. He might perhaps have surveyed the fourteen uses of the genitive absolute by the author of Hebrews to gain an idea of his proclivities, to determine whether “usually temporal” applies to him.
A larger perspective is provided by an approach known as discourse analysis. Taking this into account, we note first of all that the epistle uses time as an important theme, in particular the word “today.” The very first verse of the epistle draws a contrast between the distant past (the fathers, the prophets) and the very recent past (us, the Son), and the author builds on this idea through the whole letter. This is most evident beginning in Heb. 3:7: “Today, if you hear his voice, do not harden your hearts as in the rebellion…” He builds on “today” again in vv. 13 and 15, and repeats it in 4:7. His primary point for his readers is that their experience of the recent past, reception of the gospel of Christ, supersedes the traditions of the distant past, and demands a choice in the present: “Today.”
This is exactly what the author is saying as he begins chapter 2:
“Therefore we must pay much closer attention to what we have heard, lest we drift away from it.” (v. 1)
He then lays out again the distant past/recent past contrast, noting the superiority of the latter:
For since the message declared by angels (distant past) proved to be reliable, and every transgression or disobedience received a just retribution, how shall we escape if we neglect such a great salvation? (recent past) (vv. 2-3)
He then elaborates on the preeminence of the recent past experience:
a. “It was declared at first by the Lord,”
b. “it was attested to us by those who heard,” and
c. “God also bore witness by signs and wonders and various miracles and by gifts of the Holy Spirit distributed according to his will.”
“Angels” is easily trounced by Christ, the apostles, God himself, signs, wonders, miracles, spiritual gifts.
That all these are in the past from the perspective of the readers is not due to any hinted cessation of activity, but due to the author’s theme that what the readers have already been told requires their allegiance in the present time.
What then is the genitive absolute of verse 4 doing? I mentioned earlier how Greek uses grammatical form to signal relative prominence between predications. In this case the genitive absolute, calling attention to itself by its being in sentence final position, seems rather to serve as a prominence-enhancing device.
The author lays out the three witnesses to the gospel in stair-step fashion, with increasing prominence leading to a climax, the three demonstrating the superiority of the gospel to the Old Covenant Law: first, Christ, then the apostles, then God’s own testimony. That this is a case of increasing prominence is indicated by other grammatical features.
First the reference to Christ in v. 3a uses a very unusual structure involving yet another circumstantial participle and an infinitive, which places it on the bottom rung in terms of prominence, especially since the infinitive is passive. The phrase archen labousa laleisthai, literally means something like “taking beginning to be spoken.” Another semantic feature involved here is known as focus. At this point the message is in focus (indicated by soterias, “salvation”) and the messenger is de-focused, in this case, perhaps surprisingly, this is Christ. It is not indicating He is unimportant, but the text here takes Him out of the focus momentarily by relegating reference to him to by an agency phrase, dia tou kuriou, “through the Lord,” and this at the end of the unit.
That the next step rises in prominence is accomplished primarily by shifting to the finite verb of the sentence ebebaiothe, “was confirmed.” This too is passive, however, and the focus continues to be on the message. The apostles too appear in another agent phrase “by those who heard.”
With verse four, we reach the third step. There is now a focus shift from the message to the messenger, who is God. We have active voice rather than passive, and the verb, the genitive participle appears at the head. I am suggesting that in this case the genitive absolute construction, in its striking position following the main verb also serves to give it prominence. Beyond this, the author then piles on specifics as to the means of God’s testimony. The multiplicity is emphasized by the correlative conjunction structure te kai… kai… kai…, “both and… and… and…”
I think a better case can be made for this function than to treat it as adverbial, temporal. This is not to say that its time is not roughly simultaneous to the apostles’ confirmation of Christ’s message. I think it is, but this is not what the author is getting at. So even if these three witnesses are all in the readers’ recent past, nothing here gives the slightest hint that what was experienced by “us,” i.e. the readers, will not be experienced by others later. As Dr. Wallace himself says, there is no legitimate way to infer that it does.
To conclude, I’m afraid I cannot grant much solidity to the suggested inferences in Dr. Wallace’s article. They fail to consider context, rely on misleading statistics, distort the emphasis of the passage, and distract from the rather clear observation that the Cessationism claim remains without legitimate Scriptural support. Certainly it is not to be found in Heb. 2:3-4.