“Problematic Texts” for Definite Atonement

in the Pastoral and General Epistles

Thomas R. Schreiner

Is definite atonement actually taught in the Scriptures, or do prejudiced interpreters read it into biblical texts? I. Howard Marshall asks the right question: “Is it possible to interpret the election statements in such a way as to be consistent with the universal statements without twisting the meaning of either?”1 I will argue here that supporters of definite atonement can answer that question in the affirmative. A number of texts in the Pastoral Epistles, the Petrine Epistles, and Hebrews that speak to the issue of definite atonement will be considered. Many of the texts examined here are part of the arsenal of those who defend unlimited/general atonement. In this chapter, I will argue that (1) understanding some of these texts in a way that supports definite atonement is more persuasive exegetically and theologically; and (2) those texts which do concern God’s salvific stance to all kinds of people (1 Tim. 2:4; 4:10) or to everyone (2 Pet. 3:9) do not in fact disprove the doctrine of definite atonement—God’s desire for people to be saved and his intention to save only the elect are compatible elements in biblical soteriology.

Pastoral Epistles


As most commentators agree, a mirror reading of 1 Timothy suggests that in this epistle the apostle Paul confronts some kind of exclusivism heresy. Perhaps Paul’s opponents relied on genealogies to limit salvation to only a certain group of people, excluding from God’s saving purposes those who were notoriously sinful or those from so-called inferior backgrounds (1:4; cf. Titus 3:9).2 Paul writes to remind Timothy and the church that God’s grace is surprising: his grace reaches down and rescues all kinds of sinners, even people like Paul who seem to be beyond his saving love (1:12–17).


Paul’s reflections on his own salvation function as an important backdrop for the discussion of salvation in 1 Timothy 2:1–7, a key passage relating to definite atonement. Some contend that the emphasis on “all” precludes definite atonement.3 Paul begins by exhorting his readers to pray “for all people” (ὑπὲρ πάντων ἀνθρώπων; v. 1). Does Paul refer here to every person without exception or to every person without distinction? The immediate reference to “kings and all who are in high positions” (v. 2) suggests that various classes of people are in view.4Is such a reading of 1 Timothy 2:1–2 borne out by the subsequent verses? Praying for all is “good” and “pleasing” (v. 3), for God “desires all people to be saved and to come to the knowledge of the truth” (ὃς πάντας ἀνθρώπους θέλει σωθῆναι καὶ εἰς ἐπίγνωσιν ἀληθείας ἐλθεῖν; v. 4). The same question arising in verse 1 surfaces here again: Does “all people” (πάντας ἀνθρώπους; v. 4) refer to every person without exception or to every person without distinction? The Reformed have traditionally defended the latter option.5 Sometimes this exegesis is dismissed as special pleading and attributed to Reformed biases. Such a response is too simplistic, for there are good contextual reasons for such a reading. A focus on all people without distinction is supported by verse 7, where Paul emphasizes his apostleship and his ministry to the Gentiles: “For this I was appointed a preacher and an apostle (I am telling the truth, I am not lying), a teacher of the Gentiles in faith and truth.” Hence, there are grounds in the context for concluding that “all people” zeros in on people groups, so that Paul is reflecting on his Gentile mission. In Acts 22:15 (NIV), when Paul speaks of being a witness “to all people” (πρὸς πάντας ἀνθρώπους), he clearly does not mean all people without exception; “all” refers to the inclusion of the Gentiles in his mission (Acts 22:21).6

The parallel with Romans 3:28–30 provides further evidence that Paul thinks particularly of all people without distinction in 1 Timothy 2:4.7 Both Jews and Gentiles, according to Paul, are included within the circle of God’s saving promises. Paul contends that both are justified by faith, for the oneness of God means that there can be only one way of salvation (cf. 1 Tim. 2:5). One of the advantages of the people group interpretation is that it centers on a major theme in Pauline theology, namely, the inclusion of the Gentiles.

Such an interpretation does not seem to be special pleading, for even interpreters unsympathetic to the Reformed position detect an emphasis on Gentile inclusion in response to some kind of Jewish exclusivism (1 Tim. 1:4). For example, Marshall says, “This universalistic thrust is most probably a corrective response to an exclusive elitist understanding of salvation connected with the false teaching. . . . The context shows that the inclusion of Gentiles alongside Jews in salvation is the primary issue here.”8 And Gordon Fee remarks on verse 7, “This latter phrase in particular would seem to suggest some form of Jewish exclusivism as lying at the heart of the problem.”9

In sum, Paul reminds his readers of a fundamental truth of his gospel: God desires to save all kinds of people.10 As William Mounce says, “the universality of salvation [is] the dominant theme” in the paragraph.11 The idea of salvation is supported by the phrase “to come to the knowledge of the truth” (εἰς ἐπίγνωσιν ἀληθείας ἐλθεῖν; v. 4), which is simply another way of describing the gospel message of salvation (cf. 2 Tim. 2:25; 3:7; cf. Titus 1:1). The universal reach of salvation flows from a fundamental tenet of the OT and Judaism: there is only one God (cf. Deut. 6:4). Since there is only one God, there is only one way of salvation, for “there is one mediator between God and men, the man Christ Jesus” (εἷς καὶ μεσίτης θεοῦ καὶ ἀνθρώπων, ἄνθρωπος Χριστὸς Ἰησοῦς; 1 Tim. 2:5). God’s saving intentions are universal, including both Jews and Gentiles.

Marshall objects to the Reformed interpretation of all kinds of people, arguing that dividing groups from individuals fails, “since in the last analysis divisions between individuals and classes of humankind merge into one another.”12 But the Reformed view does not exclude individuals from God’s saving purposes, for people groups are made up of individuals. The exegetical question centers on whether Paul refers here to every person without exception or every person without distinction. We have already seen that there is strong evidence (even in Marshall) that the focus is on the salvation of individuals from different people groups. For example, in his paper, “Universal Grace and Atonement in the Pastoral Epistles,” Marshall states,

The pastor [Paul] is emphasizing that salvation is for everybody, both Jew and Gentile. . . . But it does not help the defender of limited atonement, any more than the view that “all” refers to “all kinds of people,” for what the Pastor is telling his readers to do is to pray for “both Jews and Gentiles,” not for the “the elect among Jews and Gentiles.”13

Marshall fails to see that by arguing that prayers are to be made for “Jews and Gentiles” he inadvertently affirms what he earlier denies: the Reformed position of “all kinds of people.” Moreover, Marshall actually misrepresents the Reformed view here, which is not that Paul teaches that our prayers should be limited to the elect. The Reformed position has consistently maintained that we are to pray for Jews and Gentiles, Armenians and Turks, Tutsis and Hutus, knowing that God desires to save individuals from every people group. Knowing this does not mean that we know who the elect are so that we limit our prayers to them.

The interpretation of “all without distinction” should be carried over into 1 Timothy 2:6. Here Christ is designated as the one “who gave himself as a ransom [ἀντίλυτρον] for all.”14 Clearly, we have the idea of Christ’s substitutionary sacrifice, where he gives his life as a ransom for the sake of others.15 It seems best to take the “all” (πάντων) in the same sense as we saw earlier (vv. 1, 4), meaning all kinds of people, since Paul particularly emphasizes his Gentile mission in the next verse (v. 7). Moreover, Paul most likely alludes here to Jesus’s teaching that he gave “his life as a ransom [λύτρον] for many [πολλῶν]” (Matt. 20:28; Mark 10:45), which in turn echoes Isaiah 53:11–12. As Alec Motyer demonstrates elsewhere in this volume, the referent of “many” in Isaiah 53, though it encompasses an undefined but numerous group of people, is still necessarily limited—it refers to those for whom redemption is both accomplished and applied—and therefore cannot refer to every single person.16 If these intertextual connections are correct, then Christ giving himself as a ransom for “all without exception” is ruled out.17

First Timothy 2:6 supports the notion that Christ purchased salvation for all kinds of individuals from various people groups. The verse and context say nothing about Christ being the potential ransom of everyone. The language in verse 6—“who gave himself” (ὁ δοὺς ἑαυτόν)—is a typically Pauline way of referring to the cross, and always refers to Christ’s actual self-sacrifice for believers (Rom. 8:32; Gal. 1:4; 2:20; Eph. 5:2; Titus 2:14). It stresses that Christ gave himself as a ransom so that at the cost of his death he actually purchased those who would be his people. The reason Paul can speak of Christ’s death in expansive, all-inclusive terms in 1 Timothy 2:6 is because he sees his ministry as worldwide (2:7; cf. Acts 22:15), his soteriology is universal in the right sense (2:5; cf. Rom. 3:28–30), and he is confronting an elitist heresy that was excluding certain kinds of people from God’s salvation (1 Tim. 1:4). Paul wants to make it clear: Christ died for all kinds of people, not just some elite group.18

1 TIMOTHY 4:10

Interpreters have long debated the meaning of the Pauline affirmation that God “is the Savior of all people, especially of those who believe” (ὅς ἐστιν σωτὴρ πάντων ἀνθρώπων μάλιστα πιστῶν; 1 Tim. 4:10). One aspect of the debate centers on the meaning of the word μάλιστα. The ESV translates the word “especially,” as do virtually all English translations. In 1979, however, T. C. Skeat argued that μάλιστα should be translated “namely,” or “that is.” Skeat defended his case by citing some examples from Greek papyrus letters, and then with a few NT examples. For instance, according to Skeat, when Paul asked Timothy to bring him “the books, and above all the parchments” (τὰ βιβλία μάλιστα τὰς μεμβράνας; 2 Tim. 4:13), the “parchments” define what books should be brought to him. Similarly, the “empty talkers and deceivers” (ματαιολόγοι καὶ φρεναπάται) are identified as “the circumcision party” (οἱ ἐκ τῆς περιτομῆς) using the word μάλιστα in Titus 1:10. Or, when Paul says that one should provide “for his relatives,” he defines them as “members of his household” (εἰ δέ τις τῶν ἰδίων καὶ μάλιστα οἰκείων οὐ προνοεῖ; 1 Tim. 5:8). So here in 1 Timothy 4:10, according to Skeat, the text should be translated, “God, who gives salvation to all men—that is to say, to all who believe in Him.”19 Skeat’s claim that μάλιστα means “that is” or “namely” certainly yields a coherent and plausible reading of some verses.

Nevertheless, the notion that μάλιστα means “that is” or “namely” should be rejected. Vern Poythress disputes every one of Skeat’s examples, showing that his understanding of the term is flawed in both the Greek papyri and in the NT examples.20 He shows that Skeat’s readings are either ambiguous and therefore not proven, or they are mistaken. The ambiguous texts, which could possibly support Skeat’s hypothesis, should not be introduced in favor of his interpretation. Poythress, correctly, objects that a new meaning for a word must not be accepted in ambiguous texts if an established meaning for the word makes sense in the text under consideration. He argues that the meaning “especially” or “particularly,” an elative sense of μάλιστα, fits every example. In other words, the term μάλιστα should be rendered “especially” or “particularly”; it intensifies adverbially the word it modifies.

For the sake of space we will not rehearse here the extrabiblical evidence provided by Skeat. Suffice it to say that Poythress demonstrates in every instance that Skeat’s rendering is unpersuasive. The word μάλιστα is found six times in 2–4 Maccabees and never means “that is” or “namely” (2 Macc. 8:7; 3 Macc. 5:3; 4 Macc. 3:10; 4:22; 12:9; 15:4). The two examples in Acts also should be translated “especially.” Acts 20:38 says that those who accompanied Paul to the ship were “especially sorrowful” (ὀδυνώμενοι μάλιστα) that they would not see him again. Acts 25:26 is particularly helpful. Festus, in introducing Paul to his guests, explains that he “brought him before you all, and especially before you, King Agrippa” (προήγαγον αὐτὸν ἐφ᾽ ὑμῶν καὶ μάλιστα ἐπὶ σοῦ, βασιλεῦ Ἀγρίππα; Acts 25:26). Any notion thatμάλιστα means “that is” here is clearly wrong, for the plural “you” refers to the guests, and Agrippa is distinguished from them as the special guest of the occasion.

There are some instances where Skeat’s interpretation is contextually possible. The saints who greet the Philippian believers could be identified as those who are part of Caesar’s household (Phil. 4:22). But it is much more likely that the saints and those of Caesar’s household are not coextensive. Hence, the saints with Paul greet the Philippians, and in particular or especially (μάλιστα) “those of Caesar’s household” (δὲ οἱ ἐκ τῆς Καίσαρος οἰκίας). Similarly, it fits better with the lexical meaning of μάλιστα if, in Titus 1:10, “those of the circumcision party” are a subset of the “empty talkers and deceivers.” All those of the circumcision party are empty talkers and deceivers, but there are also empty talkers and deceivers who do not belong to the circumcision group.21 Similarly, 2 Timothy 4:13 fits with whatμάλιστα means elsewhere, for it makes perfect sense to ask for books in general and then to specify that Timothy should particularly bring the parchments.

Other uses in Paul confirm that μάλιστα means “especially” or “particularly.” For instance, Paul commands the Galatians to “do good to everyone, and especially [μάλιστα] to those who are of the household of faith” (Gal. 6:10). “Everyone” is a broader category than “the household of faith,” for it includes those who are unbelievers. So, Paul admonishes the church to do good to all people but especially to fellow believers. Similarly, in Philemon 16, Paul admonishes Philemon to receive Onesimus as a brother in the Lord, adding “especially to me” (μάλιστα ἐμοί). Again, Skeat’s translation would not fit at all here. In 1 Timothy 5:8, providing for one’s own “and especially [μάλιστα] for members of [one’s] household” is naturally read as saying that the latter is a subset of the former. Those who are part of one’s household have a special priority. So too, in 1 Timothy 5:17, “elders who rule well” (οἱ καλῶς προεστῶτες πρεσβύτεροι) should receive “double honor” (διπλῆς τιμῆς), and then Paul adds, “especially those who labor in preaching and teaching” (μάλιστα οἱ κοπιῶντες ἐν λόγῳ καὶ διδασκαλίᾳ). Given the meaning of μάλιστα elsewhere, it is likely that Paul commends a subcategory of elders—those who devote themselves to the preaching and teaching of the Word.

In conclusion, then, there is little doubt that μάλιστα means “especially” instead of “that is” or “namely” in 1 Timothy 4:10. Naturally the translation “that is” would appear to fit nicely with definite atonement, for then the verse would teach that God is the Savior of all people, that is, believers. The “all people” would be defined as believers, and thus there would be no sense that God universally saves all people. Lexically, however, this interpretation is quite implausible and hence it should be rejected. The ESV translates the verse well: God “is the Savior of all people, especially of those who believe.”

Now at first glance 1 Timothy 4:10 could be interpreted to support universalism, since the verse says that God “is the Savior of all people.” But a universalist meaning is ruled out by the addition of the words “especially believers,” which are superfluous if all are saved, for it is difficult to see how believers are saved in a special way if all people without exception are saved. If universalism is true, all without exception are saved, and there is no unique salvation for believers. Furthermore, even in 1 Timothy, Paul teaches a final destruction of the impenitent, which does not fit with a universalist reading (e.g., 6:9).

But what does the verse mean if the ESV translation is accurate? The phrase “all people” (πάντων ἀνθρώπων) could be translated “all sorts of peoples,” and then the focus would be on various people groups.22 Naturally this fits well with what we have seen earlier in 1 Timothy 2:1–7 and Titus 2:11.23 However, this still begs the question of how God may be the Savior of all kinds of people, and especially of believers.

Steven Baugh proposes an interpretation that appears to solve any dilemma for a Reformed position on definite atonement. He argues that the word “Savior” here does not refer to spiritual salvation, “but to God’s gracious benefactions to all of humanity,”24 or, “to God’s care for all of humanity during our time upon earth.”25 Baugh notes many examples in Greco-Roman literature, and especially in Ephesian inscriptions, where Savior refers to the protection and preservation granted by kings, emperors, patrons, and other leaders. Paul counters the idea, according to Baugh, that those who were deceased were gods and saviors. Hence, identifying God as Savior denotes what is often called his common grace, which is granted to all people. Baugh understands the verse to say that God bestows his common grace on all people without exception. Perhaps we can think here of the provision of food, health, and the times of joy (cf. Acts 14:17). God’s goodness has been especially manifested to those who are believers, for they have been given both material and spiritual blessings.

Baugh’s interpretation solves the problem before us, for if the verse does not refer to spiritual salvation, there is no need to suggest that God secures the salvation of all people. Nevertheless, it is quite unlikely that Baugh’s interpretation is correct, for there is a crucial problem with his interpretation. One of the major themes in the Pastoral Epistles is salvation. Paul refers to both God and Christ as “Savior” (σωτήρ) and uses the verb “save” (σώζω) seven times (1 Tim. 1:15; 2:4, 15; 4:16; 2 Tim. 1:9; 4:18; Titus 3:5). God is identified as “Savior” six times in the Pastorals (1 Tim. 1:1; 2:3 4:10; Titus 1:3; 2:10; 3:4) and Christ four times (2 Tim. 1:10; Titus 1:4; 2:13; 3:6). The noun “salvation” (σωτήρια) is used twice (2 Tim. 2:10; 3:15), and the adjective “bringing salvation” (σωτήριον) once (Titus 2:11). What is striking is that there is not a single instance in the Pastorals where the salvation word group refers to anything besides spiritual salvation.26In other words, the term never means preservation, nor does it focus on material blessings. A survey of some examples will confirm this judgment.

In 1 Timothy 1:1, God as Savior is connected with the hope that belongs to believers in Christ, which makes it clear that spiritual salvation is in view. It is even clearer that spiritual salvation is intended in 1 Timothy 2:3–4, for God “our Savior” (τοῦ σωτῆρος ἡμῶν; v. 3) is the one “who desires all people to be saved” (ὃς πάντας ἀνθρώπους θέλει σωθῆναι; v. 4). Then Paul proceeds to speak of Christ as the “Mediator” (μεσίτης; v. 5), so there is no doubt that salvation from sin is the subject. A reference to spiritual salvation is evident in 1 Timothy 1:15: “Christ Jesus came into the world to save sinners” (Χριστὸς Ἰησοῦς ἦλθεν εἰς τὸν κόσμον ἁμαρτωλοὺς σῶσαι). Similarly, in 2 Timothy 1:10, Christ is identified as Savior (σωτῆρος), as the one “who abolished death and brought life and immortality to light through the gospel” (καταργήσαντος μὲν τὸν θάνατον φωτίσαντος δὲ ζωὴν καὶ ἀφθαρσίαν διὰ τοῦ εὐαγγελίου). The references to conquering death and the dawn of life through the gospel confirm a reference to spiritual salvation. In 2 Timothy 2:10, “salvation” (σωτηρίας) is linked with obtaining “eternal glory” (δόξης αἰωνίου). The Scriptures lead to “salvation through faith in Christ Jesus” (σωτηρίαν διὰ πίστεως τῆς ἐν Χριστῷ Ἰησοῦ; 2 Tim. 3:15). So too, the Lord will “save” (σώσει) Paul “into his heavenly kingdom” (εἰς τὴν βασιλείαν αὐτοῦ τὴν ἐπουράνιον; 2 Tim. 4:18).27 God and Christ are both identified as Savior (σωτῆρος) in the introduction of Titus (1:3–4), and spiritual salvation is clearly in view, since in the context Paul refers to “God’s elect” (ἐκλεκτῶν θεοῦ), “knowledge of the truth” (ἐπίγνωσιν ἀληθείας; v. 1), “eternal life” (ζωῆς αἰωνίου; v. 2), his “preaching” (κηρύγματι; v. 3), and “common faith” (κοινὴν πίστιν; v. 4). In Titus 2:10, God as “Savior” (σωτῆρος) is linked with his bringing “salvation for all people” (σωτήριος πᾶσιν ἀνθρώποις; v. 11) and “waiting for our blessed hope” (τὴν μακαρίαν ἐλπίδα; v. 13) of the coming of Christ as “God and Savior” (θεοῦ καὶ σωτῆρος). Both God and Christ are identified as Savior (σωτῆρος) in Titus 3:4–6, and this is linked with the truth that God “saved us” (ἔσωσεν ἡμᾶς; v. 5).

Lexically, then, there is little doubt that Paul refers to spiritual salvation in 1 Timothy 4:10. Surprisingly, Baugh does not consider how “salvation” and “Savior” are used elsewhere in the Pastorals, and he wrongly resorts to how the word is used in inscriptions in Ephesus instead of relying on the nearer and more important context—the Pauline usage in the Pastoral Epistles. A reference to spiritual salvation is confirmed by the context in which verse 10 appears. Paul explicitly contrasts spiritual and physical training (vv. 7–8), prizing the former over the latter. Indeed, spiritual training is paramount, for it provides benefit both “for the present life and also for the life to come” (ζωῆς τῆς νῦν καὶ τῆς μελλούσης; v. 9). The reference to “the life to come” indicates that spiritual salvation is intended.

In conclusion, Baugh’s interpretation is creative and solves the problem before us, but it fails lexically and does not account well for the meaning of “salvation” and “Savior” in the Pastoral Epistles, and therefore should be rejected.

What then is the best interpretation of 1 Timothy 4:10? We have seen thus far: (1) that the word μάλιστα means “especially”; (2) that universalism is excluded; (3) that “all people” probably focuses on people groups (both Jews and Gentiles); and (4) that “Savior” refers to spiritual salvation.

Further light may be shed on this difficult verse by seeing its parallelism with 1 Timothy 2:3–4:28

. . . God our Savior, who desires all people to be saved and to come to the knowledge of the truth (2:3–4)

. . . τοῦ σωτῆρος ἡμῶν θεοῦ, ὃς πάντας ἀνθρώπους θέλει σωθῆναι καὶ εἰς ἐπίγνωσιν ἀληθείας ἐλθεῖν

. . . the living God, who is the Savior of all people, especially of those who believe (4:10)

. . . θεῷ ζῶντι, ὅς ἐστιν σωτὴρ πάντων ἀνθρώπων μάλιστα πιστῶν

The phrase “God our Savior, who desires all people to be saved” (2:3b–4a) shares the same conceptual horizon with “the living God, who is the Savior of all people” (4:10b–c) and refers to God’s salvific desire toward all kinds of people—in this sense God avails himself as Savior to all kinds of individuals from diverse people groups. The phrase “to come to the knowledge of the truth” (2:4b) mirrors “especially . . . those who believe” (4:10d), showing that salvation is a reality only for those who come to the knowledge of the truth through faith. It seems, then, that Paul is saying here that God is potentially the Savior of all kinds of people—in that, as the living God there is no other Savior available to people—but that he is actually the Savior of only believers. The additional comment, “especially of believers,” intensifies the meaning of salvation. The possibility of God being a Savior for all kinds of people exists because there is only one living God (4:10b) and one Mediator available to people (2:5–6), but this possibility becomes a reality for those who believe. The phrase clarifies that believers are a subset of all people; they are a special category because they are actually saved.

But does such an interpretation disprove definite atonement? In the first place, this interpretation should not be confused with one that suggests two levels to the atonement: Christ dies for everyone to make them redeemable, and he dies for the elect to actually redeem them.29This introduces an unwarranted split-level into the atonement. The issue in 1 Timothy 4:10 is not two levels to the atonement, but rather the twin truths that God (the Father) is the available Savior for all kinds of people—God’s salvific stance—while at the same time being theactual Savior for only those who believe (in Christ).

Secondly, 1 Timothy 4:10 illustrates that definite atonement may be affirmed alongside other biblical truths, such as God’s salvific stance to the world and the possibility for people to be saved if they believe in Christ. Those who hold to a definite intention in the atonement to save only the elect also believe that God desires people to be saved (1 Tim. 2:3–4; cf. Ezek. 18:32), that he is available as Savior to all people (1 Tim. 4:10), that Christ’s death is sufficient for the salvation of every person,30 and that all are invited to be saved on the basis of Christ’s death for sinners (1 Tim. 1:15). But it is a non sequitur to suggest that affirming any of these biblical truths somehow negates the truth that Christ intended to die only for his elect, actually paying for their sins alone. In biblical soteriology, these theological elements sit side by side.

TITUS 2:11–14

Another text that pertains to definite atonement in the Pastorals is Titus 2:11–14. Verse 11 is particularly striking: “For the grace of God has appeared, bringing salvation for all people” (Ἐπεφάνη γὰρ ἡ χάρις τοῦ θεοῦ σωτήριος πᾶσιν ἀνθρώποις). We are again faced with the issue that has occupied us in 1 Timothy. Some maintain that “all people” (πᾶσιν ἀνθρώποις) refers to all people without exception, but it is more likely that Paul again refers to all people without distinction. A good case can be made for such a judgment, because Paul refers to people from various groups earlier in chapter 2: older men (v. 2), older women (vv. 3–4), younger women (vv. 4–5), younger men (v. 6), and slaves (vv. 9–10). Indeed, verse 14 focuses particularly on Christ’s redeeming work for believers: Christ “gave himself for us [ὑπὲρ ἡμῶν] to redeem us [λυτρώσηται ἡμᾶς].” The repeated use of the first person plural pronoun “us” (ἡμῶν, ἡμᾶς) in the text (2:12, 14) points to Christ securing salvation for his own. Furthermore, the ἵνα clause shows that Christ’s intention was not merely to make salvation possible for everyone, but to actually redeem (λυτρώσηται) and purify (καθαρίσῃ) a special people for himself (ἑαυτῷ λαὸν περιούσιον).

Petrine Epistles


Space precludes an exhaustive assessment of Peter’s soteriology in his epistles,31 but a quick survey reveals that they are rich in the theology of election and atonement (e.g., 1 Pet. 1:1–2, 8–9, 20; 2:24; 3:18).32 For the purposes of this chapter, however, my focus concerns two Petrine texts that are often adduced to refute definite atonement: 2 Peter 2:1 and 3:9.

2 PETER 2:1

It seems to some as if 2 Peter 2:1 presents a case that is contrary to definite atonement, for in speaking of the false teachers, who initially embraced the gospel but have now denied it, Peter says that they are “denying the Master who bought them.”33 What is quite striking is that Peter says that Christ “bought them” (ἀγοράσαντα αὐτούς). What Peter means here has been interpreted in different ways. Some argue that the buying here is non-soteriological, and hence Peter does not teach that Christ redeemed the false teachers.34 The problem of Christ actually purchasing believers who then lose the benefit of being purchased is thereby avoided. But this interpretation faces a severe lexical problem. We have no instance in the NT where the ἀγοράζω word group, when it is associated with the death of Christ, has a non-soteriological meaning (cf. 1 Cor. 6:20; 7:23; Gal. 3:13; 4:5). So, this interpretation looks like special pleading in which the word “bought” is redefined to salvage the theology of definite atonement. Gary D. Long defends another non-soteriological view. He argues that δεσπότηςhere refers to Christ as Creator and that ἀγοράζω is a creation term as well, referring to Christ’s ownership of the false teachers.35 But Long’s view fails for the same reason as the view examined above, for we have already seen that the ἀγοράζω word group is soteriological in the NT.36

Another possibility is that the word “bought” bears its usual meaning, but those who were bought or redeemed fell away from the faith. The false teachers were truly redeemed by the blood of Christ but they apostatized and denied the faith that they had at first embraced. This is another way of saying, of course, that they lost or abandoned their salvation.37 On this reading, some of those whom Christ has redeemed or purchased end up being damned. The apostasy view has the advantage of being a straightforward and clear reading of the text. Some of those whom Christ redeemed have fallen away and denied the faith. Space is lacking to interact in detail, either exegetically or theologically, with the notion that some of those who are redeemed may end up eternally damned.38 I would argue that there are many texts which teach that those who truly belong to the Lord will never finally and ultimately fall away, since the Lord has promised to keep them (see e.g., John 10:28–29; Rom. 8:28–39; 1 Cor. 1:8–9; Phil. 1:6; 1 Thess. 5:23–24). Hence, the loss-of-salvation view should be rejected.

D. W. Kennard proposes another solution to the text before us.39 The term “bought,” says Kennard, is soteriological. The false teachers, therefore, were genuinely bought or redeemed by Christ. Kennard, however, departs from both standard Arminian and Reformed views in explaining the nature of redemption here, for he maintains that some of those who are redeemed will not be saved on the final day. At first glance one might conclude that this interpretation fits with Arminianism since some of those who are truly redeemed will lose their redemption, and hence will not be saved on the day of judgment. Kennard, nevertheless, introduces a wrinkle that distinguishes him from classical Arminianism, for on his scheme all the elect will certainly be saved and will never lose their elect status. According to Kennard, however, some of those who are redeemed are not elect.

How should Kennard’s proposal be assessed? It would take us too far afield to consider his proposal in detail, for we would need to investigate the nature of redemption and election elsewhere in the NT. Suffice it to say that his reading, which separates election from redemption, is unpersuasive and lacks exegetical and theological support from the remainder of the NT. Traditional Arminian and Reformed scholars offer more plausible readings when they posit, respectively, that either those who are elect and redeemed may apostatize or that those who are elect and redeemed will surely be kept from apostasy by God himself.

Still another possible reading has been proposed. The term “bought” here refers to what Andrew Chang calls “spiritual redemption.”40The atonement is unlimited in nature; the problem with the false teachers is their refusal to accept the salvation purchased for them. This view must be distinguished from the “loss-of-salvation” notion presented above, for Chang insists that no true believer can apostatize. The Arminian interpretation says that some were truly redeemed but repudiated their salvation. But Chang maintains that Peter describes the false teachers as “bought” in terms of potentiality. Theologically, this interpretation ends up saying that Christ purchased all potentially, but the purchasing does not take effect unless someone believes.

Chang’s interpretation, though it may seem appealing at first glance, should be rejected. When we approach a text, it is vital to read it in context. We must attend to what the text we are investigating is trying to do, so that we read it on its own terms. Chang’s interpretation fails to convince because he separates what Peter says about the false teachers being redeemed by Christ from what Peter says about their falling away, in 2 Peter 2:20–22. The false teachers are described as those who “have escaped the defilements of the world through the knowledge of our Lord and Savior Jesus Christ” (v. 20).41 Verse 21 says that they “have known the way of righteousness.” So, it is rather astonishing that Chang says, “The text gives no evidence that these false teachers professed to be believers.”42 Peter remarks that after having escaped, they have now been “entangled” and “overcome” (v. 20), so that “the last state has become worse for them than the first” (v. 20). They have “turn[ed] back from the holy commandment delivered to them” (v. 21). So, they are like unclean dogs and pigs, who have revisited their filth. Peter describes the false teachers as being purchased by Christ (v. 1), as knowing Jesus as Lord and Savior (v. 20), and as knowing the righteous way (v. 21). It is precisely here where it is evident that Chang’s solution does not work, for Peter is not saying that the false teachers potentially knew Christ as Lord and Savior or that they potentially knew the righteous way. It is evident from Peter’s language that the false teachers gave every indication initially that they were truly Christians. Chang’s view lacks inner coherence and consistency, for he fails to integrate what Peter says about the false teachers being bought by Christ (v. 1) with their knowing Christ as Lord and Savior (v. 20) and knowing the way of righteousness (v. 21).

Is there a reading that treats this text plausibly, and consistently interprets what Peter says about the false teachers in both verse 1 and verses 20–22? I suggest there is: Peter’s language is phenomenological. In other words, it appeared as if the Lord had purchased the false teachers with his blood (v. 1), though they actually did not truly belong to the Lord.43 Similarly, the false teachers gave every appearance of knowing Jesus Christ as Lord and Savior (v. 20) and appeared to have known the righteous saving way (v. 21).44 Such an interpretation is to be preferred to Chang’s reading, for the same interpretation is proposed for verse 1 and verses 20–21. In both instances a phenomenological reading makes good sense of the text, whereas it does not work to speak of a potential redemption (v. 1) and a potential knowing of Christ (vv. 20–21), for Peter says that they knew the Lord, and hence he does not refer to potentiality in verses 20–21. The issue is whether the language of being bought by Christ and knowing the Lord is plausibly interpreted as phenomenological.

Why would Peter use phenomenological language if the false teachers were not truly saved? Is this an artificial interpretation introduced to support a theological bias? I have already said that the Arminian reading of the text is straightforward and clear. One can understand why it has appealed to so many commentators throughout history. However, it is better to say that the false teachers gave every appearance of being saved. They seemed to be part of the redeemed community, but their apostasy demonstrated that they never truly belonged to God. The words of 1 John 2:19 fit them: “They went out from us, but they were not of us; for if they had been of us, they would have continued with us. But they went out, that it might become plain that they all are not of us.” Similarly, Jesus said about those who prophesied in his name, exorcised demons, and performed miracles, but who lived lawless lives, “I never knew you” (Matt. 7:23). He does not say that he knew them once but that he does so no longer. On the contrary, they were never truly members of the people of God, yet for a time they gave the impression of being so. There are other texts which teach that some who truly appeared to be believers later turned out to have spurious faith (Mark 4:1–20; 1 Cor. 11:19; 2 Tim. 2:19).45 Furthermore, Peter’s use of phenomenological language makes sense, for the false teachers were vitally involved in the church. It was not as if outsiders who never claimed to be Christians arrived and began to propagate teachings contrary to the gospel. On the contrary, the false teachers were insiders who departed from what they were first taught. Hence, Peter underscores the gravity of what occurred. Those who were fomenting the false way were, so to speak, “Christians.” They were to all appearances “bought” by Christ (2 Pet. 2:1) and seemed to “know” him as Lord and Savior (v. 20). Peter is not claiming that they were actually Christians, that they were truly redeemed (v. 1), or that they truly knew Jesus as Lord and Savior (v. 20), but that they gave every reason initially for observers to think that such was the case. Their subsequent departure showed that they were actually dogs and pigs (v. 22). In other words, they were never truly changed, and thus eventually they revealed their true nature.

To sum up, 2 Peter 2:1 does not falsify definite atonement, for Peter does not intend to teach that Christ actually or potentially redeemed the false teachers. Instead, he uses phenomenological language, which is the same way we should interpret the language of their knowing Christ as Lord and Savior (v. 20). The false teachers initially gave every impression of being believers, and thus in turn appeared to have been “bought” (in a soteriological sense) by Christ. Hence, their subsequent defection was all the more surprising.

A right understanding of 2 Peter 2:1 actually supports definite atonement, since Christ did not actually buy these false teachers—for if he had, they would have persevered. Definite atonement refers not only to the intended target of the atonement—namely, the elect—but also to its efficacy: the atonement achieves its purpose, full and final salvation for the elect. What some fail to grasp in using 2 Peter 2:1 in support of a general atonement46 is that to affirm general atonement here is to compromise the doctrine of the perseverance of the saints. For we have seen in 2 Peter 2 that what Peter teaches about the atonement (v. 1) cannot be separated from what he teaches about perseverance (vv. 20–22). No doctrine is an island, and to suggest general atonement in this verse is to distort the doctrine of Christian perseverance.47 Therefore, to say that Christ died for the false teachers phenomenologically fits both exegetically and theologically.

2 PETER 3:9

Another verse that plays a significant role in the discussion of definite atonement is 2 Peter 3:9. God is “not wishing that any should perish, but that all should reach repentance” (μὴ βουλόμενός τινας ἀπολέσθαι ἀλλὰ πάντας εἰς μετάνοιαν χωρῆσαι). Here Peter explains that God’s patience provides the reason why Jesus’s coming is delayed. The reason for his patience is then explained: he does not want any to perish but all to repent. The idea that God is patient so that people will repent is common in the Scriptures (Joel 2:12–13; Rom. 2:4). God’s slowness “to anger” is a refrain repeated often in the OT (Ex. 34:6; Num. 14:18; Neh. 9:17; Ps. 86:15; 145:8; Joel 2:13; Jonah 4:2; Nah. 1:3), but he will not delay his wrath forever.

We should note at the outset that perishing (ἀπολέσθαι) refers to eternal judgment, as is typical with the term. Repentance (μετάνοιαν), correspondingly, involves the repentance that is necessary for eternal life. Peter does not merely discuss rewards that some will receive if they live faithfully. He directs his attention to whether people will be saved from God’s wrath. We must also ask who is in view when he speaks of “any” (τινας) perishing and “all” (πάντας) coming to repentance. Notice that the verse says “patient with you” (μακροθυμεῖ εἰς ὑμᾶς). The “any” and “all” in the verse may be an expansion of “you” (ὑμᾶς) earlier in the verse. Peter does not reflect, according to one interpretation, on the fate of all people in the world without exception. He considers those in the church who have wavered under the influence of the false teachers. God desires that every one of them will repent.48

A restrictive meaning of “you” is certainly possible. But it seems more probable that the words “any” and “you” refer to God’s desire for all without exception to be saved. John Murray rightly argues that there is no definite reference to the elect in the context, that the call to repentance suggests that some of those addressed might perish if they fail to repent, and hence Peter indiscriminately summons all to repent.49

It is evident, of course, that not all are saved. So, how do we explain a desire of God that is frustrated in part? Theologians have often and rightly appealed here to two different senses in God’s will: there is a decretive will of God and a permissive will of God. God desires the salvation of all in one sense, but he does not ultimately ordain and decree that all will be saved. Is there a contradiction, though, in saying that God desires the salvation of all but decrees or determines the salvation of only some? Positing a contradiction is unconvincing, for the Scriptures teach us that there is “complexity” in the divine will.50 For instance, in Romans 9, Paul explicitly affirms God’s decretive will to elect some (Jacob and not Esau), and yet in 10:21 God stretches his hands out to all Israel in invitation because he longs for them to be saved. The two-sided dimension of God’s will is also expressed in the apostle’s ministry. In 2 Timothy 2:10, Paul says he endures all things for the sake of the elect, yet in 1 Corinthians 9:22 he becomes all things to all people so that he might save some. The “complexity” in the divine will is therefore apparent.51

If the interpretation proposed here is correct, 2 Peter 3:9 should be understood to teach that God desires the salvation of everyone. Nonetheless, it is clear from many texts that he decrees the salvation of only some. The notion that Christ died to secure the salvation of some and actually paid for the sins of those whom he has chosen fits with divine election and with the application of the Spirit’s work to the hearts of believers. The Father, Son, and Spirit work together in securing the redemption of God’s people (cf. 1 Pet. 1:1–2). From eternity past God decreed that Christ’s death would be effective for the elect. At the same time, sinners are indiscriminately offered full forgiveness because God desires all to be saved.


The main text in Hebrews that relates to definite atonement is Hebrews 2:9, where the author says that Jesus suffered “by the grace of God” (χάριτι θεοῦ) so that “he might taste death for everyone” (ὑπὲρ παντὸς γεύσηται θανάτου). Understandably this text has often been adduced to support unlimited atonement. I will argue, however, that such a reading of the text, though superficially attractive, does not fit well with the context of Hebrews 2.

Before addressing the meaning of Hebrews 2:9, a quick survey of Hebrews relative to the atonement is fitting. In Hebrews, Jesus is the Melchizedekian Priest who, in fulfillment of Psalm 110:1, “sat down at the right hand of the Majesty on high” after accomplishing “purification of sins” (1:3). Jesus, in contrast to the Levitical priesthood, brought “perfection,” in that believers now “draw near to God” (7:19) through his sacrifice. His sacrifice is permanently effective, since he intercedes for believers on the basis of his death as the one who lives and reigns forever (7:24–25). The relationship between Jesus’s death and his intercession is crucial. Clearly, Jesus’s intercession as the Risen One is invariably effective since he intercedes on the basis of his death (cf. Rom. 8:31–34). But it would be illegitimate to posit a separation between his death and intercession. In other words, Jesus intercedes specially and exclusively for those for whom he died. Just as he does not intercede for all, so in the same way he died in a unique sense for those whom he came to save, pleading on the basis of his death for their salvation.

The author of Hebrews desired his readers to be full of assurance. Hence, he reminds them that Christ’s blood cleanses their consciences (Heb. 9:14). Christ’s sacrifice is the final and definitive sacrifice (9:25–28), and hence no further sacrifice is needed. Christ has effectively borne “the sins of many” (9:28). His one sacrifice renders the need for other sacrifices superfluous (10:1–4). Believers are “sanctified through the offering of the body of Jesus Christ once for all” (10:10; cf. 10:14). Since Christ’s work on the cross is complete, he sits at God’s right hand (10:12). Sin has been completely and decisively forgiven at the cross of Christ (10:14).

The texts on intercession and sanctification point to the truth that Christ specially died for those who are his own. Nevertheless, Hebrews 2:9 could easily be understood as pointing in the other direction, since it says that Jesus tasted death for all, and Psalm 8 presumably includes a reference to every human being (cf. Heb. 2:5–8).52

Yet when we actually examine the context of Hebrews 2, we find evidence suggesting that the death Jesus tasted “for everyone” (ὑπὲρ παντός) does not, in this context, refer to everyone without exception but to everyone without distinction.53 First, in verses 5–8, though the author refers to human beings in general, he does not put any stress on all human beings without exception. Instead, the author focuses on Jesus Christ and teaches that only those who belong to him will enjoy the rule over all things described in Psalm 8. Second, verse 10 speaks of “bringing many sons to glory” (πολλοὺς υἱοὺς εἰς δόξαν ἀγαγόντα). Jesus’s suffering was effective in its design and purpose, in that it actually brought “sons to glory.” The focus clearly rests on what Jesus effectively accomplished through his death. Third, those redeemed are described as “brothers” (ἀδελφούς) of Jesus (vv. 11–12).54 Those who are the beneficiaries of Jesus’ death are identified as members of his family. Hence, the author does not call attention to the benefit of Jesus’s death for all people in general but to the advantage that exists for those who are part of his family. Fourth, the particularity in Jesus’s family is even clearer in verse 13, where the author, in citing Isaiah 8:18, depicts Jesus as saying, “Behold, I and the children God has given me” (ἰδοὺ ἐγὼ καὶ τὰ παιδία ἅ μοι ἔδωκεν ὁ θεός). Not any or all children are in view here, but specific children—the children God has given to Jesus. It seems, then, that Jesus’s brothers are equivalent to the children God gave him. Jesus suffered to bring these to glory, suggesting that his death “for everyone” in context refers to those brothers whom God had ordained to be part of his family. Fifth, in verse 16 the author of Hebrews remarks that Jesus does not help angels, “but he helps the offspring of Abraham.” The phrase “offspring of Abraham” (σπέρματος Ἀβραάμ) is most interesting. If the author had a general or unlimited atonement in view, we would expect a reference to the “offspring of Adam” or “the sons of Adam.” Such a designation would emphasize the universality of Jesus’s work for all human beings. But that is not the purpose of the author of Hebrews here. He focuses on the “offspring of Abraham,” so that the emphasis is on God’s chosen people—the children of Abraham. As we see elsewhere in the NT, the church of Jesus Christ is considered to be the seed of Abraham (cf. Gal. 3:6–9).55 Many readers may interpret the text quickly and be guilty of thinking that the “offspring of Abraham” are equivalent to the “offspring of Adam.” Clearly, the focus is not on the undifferentiated love of Christ but on his particular concern for the chosen seed of Abraham.

When we place this description of Abraham’s offspring with the emphasis on the children God gave to Jesus and the use of the word “brothers,” we have significant evidence that Jesus’s death “for everyone” (v. 9) is particular rather than general. Hence, it supports definite atonement rather than general atonement. All of this fits with verse 17, which speaks of Jesus’s High Priestly ministry “to make propitiation for the sins of the people” (εἰς τὸ ἱλάσκεσθαι τὰς ἁμαρτίας τοῦ λαοῦ). Given the focus on God’s elect and Jesus’s family in the context, it seems fair to conclude that here the emphasis is on the actual satisfaction accomplished in Jesus’s death for those who would be part of his family.56

To conclude, though Hebrews 2:9 may on first glance support general atonement, a closer look at the context suggests that definite atonement is in view.


This chapter has concentrated on texts that are often cited as disproving definite atonement. In the Pastorals, 1 Timothy 2:1–7, 4:10, and Titus 2:11 focus on salvation being accomplished for all without distinction, both Jews and Gentiles. God’s saving purposes are not restricted to the Jews but extend to the entire world. Furthermore, the salvation Christ has accomplished is effective; he has truly ransomed some to be saved (1 Tim. 2:6; Titus 2:11, 14). He has not merely made salvation possible; he has actually saved those whom he has chosen.

Second Peter 2:1, which speaks of Jesus’s redemption of the false teachers, is often cited in support of general atonement. I have attempted to show, however, that when we compare 2:1 with 2:20–22, the language of redemption is phenomenological. The false teachers appeared to be believers because of their initial embrace of the Christian faith. Their later defection showed that they were not true believers and therefore were not truly ransomed by Christ. Hence, 2 Peter 2:1 does not support general atonement, and to argue that it does is potentially to compromise Christian perseverance. Second Peter 3:9, which speaks of God’s desire for all to repent, should be interpreted as expressing God’s desired will, but God’s will of desire does not negate the fact that he has decreed that only some will be saved. We have seen in this chapter that we must distinguish between God’s desired will (his desire for all to be saved) and his decretive will (his determination that only some will be saved).

Finally, Hebrews 2:9 is regularly cited in defense of general atonement, since it speaks of Jesus’s death “for everyone.” When we consider Hebrews as a whole, the author emphasizes the effectiveness of Jesus’s death, especially in tying together Jesus’s intercession with his atoning sacrifice. Furthermore, there are significant indications in the context of Hebrews 2 that “everyone” refers to God’s chosen people, for the author speaks of the sons who are brought to glory (v. 10), of Jesus’s brothers (vv. 11–12), of the children that God gave to Jesus (v. 13), and of Abraham’s offspring (v. 16). In context, the focus is on Abraham’s family—the chosen people of God—which rules out a general atonement. Jesus’s propitiation (v. 17), then, is specifically for his people.

1 I. Howard Marshall, “Universal Grace and Atonement in the Pastoral Epistles,” in The Grace of God and the Will of Man, ed. Clark H. Pinnock (Minneapolis: Bethany, 1995), 53.

2 For a full analysis of the false teaching that Paul addresses in the Pastoral Epistles, see George W. Knight III, The Pastoral Epistles: A Commentary on the Greek Text, NIGTC (Grand Rapids, MI: Eerdmans, 1992), 10–12; I. Howard Marshall, A Critical and Exegetical Commentary on the Pastoral Epistles, ICC (Edinburgh: T. & T. Clark, 1999), 44–51; and Philip H. Towner, The Letters to Timothy and Titus, NICNT (Grand Rapids, MI: Eerdmans, 2006), 41–50. Gordon D. Fee, 1 and 2 Timothy, Titus, NIBC (Peabody, MA: Hendrickson, 1984), 64, writes, “The concern [in 1 Timothy 2:3–4] is simply with the universal scope of the gospel over against some form of heretical exclusivism and narrowness.”

3 See, for example, Marshall, “Universal Grace and Atonement,” 62–63; and Robert P. Lightner, The Death Christ Died: A Biblical Case for Unlimited Atonement, rev. ed. (Grand Rapids, MI: Kregel, 1998), 62–73.

4 So Knight, Pastoral Epistles, 115.

5 John Calvin, Institutes of the Christian Religion, ed. John T. McNeill, trans. Ford Lewis Battles (Philadelphia: Westminster, 1960), 3.24.16; John Owen, The Death of Death in the Death of Christ (Carlisle, PA: Banner of Truth, 1995), 233–35; and Knight, Pastoral Epistles, 119.

6 If “world” in 1 Timothy 3:16 refers to human beings, the term refers to every person without distinction, not every person without exception, for it is obvious that many in the world did not believe.

7 Cf. Romans 11:32, where “all” embraces Jew and Gentile, but not every person (cf. Gal. 3:28; Col. 3:11).

8 Marshall, Pastoral Epistles, 420, 427. In his comment on 1 Timothy 2:4, Marshall says, “the emphasis on ‘all’ is presumably directed at the false teaching in some way” (425).

9 Fee, 1 and 2 Timothy, Titus, 67.

10 The focus on all kinds of people ensures that whatever gender, class, economic status, social standing, or moral history, no one is excluded from God’s salvation. The “all without distinction” position is an expansive, all-inclusive one, and should not be understood otherwise.

11 William D. Mounce, Pastoral Epistles, WBC (Nashville: Thomas Nelson, 2000), 78.

12 Marshall, Pastoral Epistles, 427.

13 Marshall, “Universal Grace and Atonement in the Pastoral Epistles,” 63.

14 Leon Morris, The Apostolic Preaching of the Cross, 3rd rev. ed. (Grand Rapids, MI: Eerdmans, 1965), 51, renders ἀντίλυτρον as “substitute-ransom.”

15 Cf. Marshall, Pastoral Epistles, 432; Mounce, Pastoral Epistles, 89–90.

16 See J. Alec Motyer, “‘Stricken for the Transgression of My People’: The Atoning Work of Isaiah’s Suffering Servant,” chapter 10 in this volume.

17 Hence, the major thesis of Gary L. Shultz, Jr., “A Biblical and Theological Defense of a Multi-Intentioned View of the Extent of the Atonement” (PhD diss., The Southern Baptist Theological Seminary, 2008), that Christ actually paid for the sins of all people without exception should be rejected.

18 Some could say that Jesus is actually the ransom of all and opt for universalism, but as I point out below in the discussion on 1 Timothy 4:10, there are serious problems with a universalist reading.

19 T. C. Skeat, “‘Especially the Parchments’: A Note on 2 Timothy iv. 13,” JTS 30 (1979): 174. R. A. Campbell, “KAI MALISTA OIKEIWN—A New Look at 1 Timothy 5:8,” NTS 41 (1995): 157–60, has added support to Skeat’s position. So also Knight, Pastoral Epistles, 203–204.

20 Vern S. Poythress, “The Meaning of μάλιστα in 2 Timothy 4:13 and Related Verses,” JTS 53 (2002): 523–32.

21 Hong Bom Kim, “The Interpretation of μάλιστα in 1 Timothy 5:17,” Novum Testamentum 46 (2004): 360–68, shows that μάλιστα never means “that is” or “namely” in the Pastoral Epistles, and that the translation “especially” is correct. Surprisingly, Kim shows no awareness of Poythress’s article on the subject.

22 So, for example, Louis Berkhof, Systematic Theology, 4th ed. (Grand Rapids, MI: Eerdmans, 1941), 396–97; and Knight, Pastoral Epistles, 203.

23 Steven M. Baugh, “‘Savior of All People’: 1 Tim 4:10 in Context,” WTJ 54 (1992): 333. Though Reformed, Baugh rejects this interpretation here, but he embraces it in 1 Timothy 2:4.

24 Ibid., 331. So also John Calvin, Commentaries on the Epistles to Timothy, Titus, and Philemon, trans. William Pringle (repr., Grand Rapids, MI: Baker, 2005), 112.

25 Baugh, “Savior of All People: 1 Tim 4:10 in Context,” 333.

26 Fee, 1 and 2 Timothy, Titus, 110, rightly says that such an interpretation of Savior is “found nowhere else in the NT.” So also Knight, Pastoral Epistles, 203; and Shultz, “Multi-Intentioned View of the Extent of the Atonement,” 138–39.

27 Scholars dispute the meaning of “save” in 1 Timothy 2:15 and 4:16, but spiritual salvation is likely intended in these instances as well.

28 We should also keep in mind the context of Jewish exclusivism (1 Tim. 1:4), which Paul was addressing.

29 See, for example, D. Broughton Knox, “Some Aspects of the Atonement,” in The Doctrine of God, vol. 1 of D. Broughton Knox, Selected Works (3 vols.), ed. Tony Payne (Kingsford, NSW: Matthias Media, 2000), 260–66.

30 The sufficiency of Christ’s death is a statement of its intrinsic value unrelated to its design.

31 I am assuming here that 1 and 2 Peter were written by the apostle Peter. Second Peter is particularly controversial. For a defense of Petrine authorship, see Thomas R. Schreiner, 1 and 2 Peter and Jude, New American Commentary (Nashville: B&H Academic, 2003), 255–76.

32 For a helpful treatment on 1 Peter, see Martin Williams, The Doctrine of Salvation in the First Letter of Peter, Society for New Testament Studies Monograph Series 149 (Cambridge: Cambridge University Press, 2010).

33 For example, R. C. H. Lenski, The Interpretation of the Epistles of St. Peter, St. John and St. Jude (Minneapolis: Augsburg, 1966), 305, “Here we have an adequate answer to Calvin’s limited atonement: the Sovereign, Christ, bought with his blood not only the elect but also those who go to perdition.”

34 Wayne Grudem, Systematic Theology: An Introduction to Biblical Doctrine (Grand Rapids, MI: Zondervan, 1994), 600; Owen, Death of Death, 250–52, emphasizes the non-soteriological solution, but he also recognizes that the language may be phenomenological.

35 Gary D. Long, Definite Atonement (Phillipsburg, NJ: P&R, 1977), 67–79. Like Owen, Long acknowledges the possibility of the phenomenological view. Cf. also Baugh, “‘Savior of All People’: 1 Tim 4:10 in Context,” 331; and Calvin, Epistles to Timothy, Titus, and Philemon, 112.

36 For criticisms of Long, see Andrew D. Chang, “Second Peter 2:1 and the Extent of the Atonement,” BSac 142 (1985): 52–56.

37 So, for example, I. Howard Marshall, Kept by the Power of God: A Study of Perseverance and Falling Away (Minneapolis: Bethany, 1969), 169–70.

38 See Thomas R. Schreiner and Ardel B. Caneday, The Race Set Before Us: A Biblical Theology of Perseverance and Assurance (Downers Grove, IL: InterVarsity Press, 2001).

39 D. W. Kennard, “Petrine Redemption: Its Meaning and Extent,” JETS 39 (1987): 399–405.

40 Chang, “Second Peter 2:1 and the Extent of the Atonement,” 60.

41 Actually, what Peter says here is true both of the false teachers and of their “converts” who have also fallen away. In defense of this view, see Schreiner, 1 and 2 Peter and Jude, 360–61.

42 Chang, “Second Peter 2:1 and the Extent of the Atonement,” 56.

43 Ibid., 60, dismisses this view, which he identifies as the “Christian charity view,” saying that “the text gives no support to this view.” But he fails to see that verses 20–22 do support this view when these verses are integrated with verse 1. Indeed, the latter must not be segregated from the former, for both texts refer to the false teachers.

44 Shultz, “Multi-Intentioned View of the Extent of the Atonement,” 150 n. 180, contradicts himself in his exposition of 2 Peter 2. When referring to verse 1 and the notion that the false teachers were professing believers, he says, “There is no support for this view in the text, and there is good reason to believe that the false teachers were not professing believers.” Shortly thereafter he says, “the false teachers are not apostate Christians or former Christians who have lost their salvation” (151). But he later says about verses 20–22, “These false teachers are unbelievers who once made false professions of faith without ever experiencing regeneration” (182). Contrary to Shultz, the false teachers were “apostate Christians,” in the sense that they had fallen away from their earlier profession of faith. Shultz, like so many, fails to consider the role of verses 20–22 and what it says about the false teachers in his comments on verse 1. Hence, his dogmatic statement about there being no support for the phenomenological interpretation is false and contradicted by his own words, for if one believes that the false teachers had not lost their salvation (as Shultz does), they had at the very least renounced the profession of faith they had made previously.

45 See here D. A. Carson, “Reflections on Assurance,” in Still Sovereign: Contemporary Perspectives on Election, Foreknowledge, and Grace, ed. Thomas R. Schreiner and Bruce A. Ware (Grand Rapids, MI: Baker, 2000), 260–69, where he presents a very persuasive argument for a category of people in Scripture with spurious faith.

46 For example, Knox, “Some Aspects of the Atonement,” 263; and Mark Driscoll and Gerry Breshears, Death by Love: Letters from the Cross (Wheaton, IL: Crossway, 2008), 172.

47 Or, to avoid this, proponents must revert to the language of “potentiality,” which, as we have seen, lacks coherence in the wider context.

48 So Owen, Death of Death, 236–37.

49 John Murray, “The Free Offer of the Gospel,” in Collected Writings of John Murray. Volume 4: Studies in Theology (Edinburgh: Banner of Truth, 1982), 129–30.

50 John Calvin, Commentaries on the Catholic Epistles, ed. John Owen, trans. John Owen (repr., Grand Rapids, MI: Eerdmans, 1948), 22:419–20, defended the notion that God’s will is “complex.” Calvin says, “But it may be asked, If God wishes none to perish, why is it that so many do perish? To this my answer is, that no mention is here made of the hidden purpose of God, to which the reprobate are doomed to their own ruin, but only of his will as made known to us in the gospel. For God there stretches forth his hand without a difference to all, but lays hold only of those, to lead to them to himself, whom he has chosen before the foundation of the world” (420). It is important to qualify that God’s will is only “complex” in how it appears to us. Calvin, again, is helpful here: “God’s will is one and simple in him” but it “appears manifold to us on account of our mental incapacity” (Institutes, 1.18.3). We need to refer the “complexity” to our perception and not to divine volition per se.

51 The “complexity” in God’s will does not depend upon positing a distinction between θέλω and βούλομαι, as if the latter term refers to God’s decreed will and the former to his preference. See especially the pointed comments of Marshall, “Universal Grace and Atonement in the Pastoral Epistles,” 55–57. Cf. also Mounce,Pastoral Epistles, 86. But against Marshall, the distinction between God’s decreed and desired will rests on a larger perspective than the individual reading of particular words, and hence is still a legitimate theological conclusion.

52 So Lightner, The Death Christ Died, 71–72; and Shultz, “Multi-Intentioned View of the Extent of the Atonement,” 144.

53 Rightly, Owen, Death of Death, 238; and John Murray, Redemption Accomplished and Applied (Grand Rapids, MI: Eerdmans, 1955), 61.

54 William L. Lane, Hebrews 1–8, WBC (Waco, TX: Word, 1991), 59, says that they are part of the “covenantal family.”

55 Cf. here Harold W. Attridge, The Epistle to the Hebrews, Hermeneia (Philadelphia: Fortress, 1989), 94; and Philip E. Hughes, A Commentary on the Epistle to the Hebrews (Grand Rapids, MI: Eerdmans, 1977), 119.

56 In support of interpreting ἱλαστήριον as “propitiation” here, see Lane, Hebrews 1–8, 66, and Hughes, Hebrews, 121–23; contra Attridge, Hebrews, 96 n. 192, who argues for “expiation.” Shultz, “Multi-Intentioned View of the Extent of the Atonement,” 144, says that all things cannot be subjected to Jesus if he did not pay for the sins of all, but such a theological deduction is not warranted by the argument of Hebrews 2.


Excerpt from From Heaven He Came and Sought Her: Definite Atonement in Historical, Biblical, Theological, and Pastoral Perspective by David and jonathan Gibson,

posted with permission.