The question for this section is whether Harnack really keeps the Good Father and the Creator apart, or whether, having distinguished them historically, he finally absorbs the distinction into a single theological framework.
To His Credit
The answer is two-part, and the first part is to Harnack’s credit.
Formally he preserves the distinction with more care than almost anyone before him, and he does it with argument rather than assertion. Against Bauer’s attempt to make the Creator simply bad, he observes on his tenth page that Marcion used δίκαιος, νόμος and even ἀγαθός in a twofold sense, and that he “sharply distinguished between κακός and πονηρός”, two Greek words for badness that Bauer’s reading has to treat as interchangeable. He builds the philology out in three footnotes on his thirteenth page: Marcion left Romans 7:12 standing, “the law is holy and the commandment holy and righteous and good,” which he would have had to alter had he thought the lawgiver bad at the root; he cites Theodoret for δίκαιος καὶ πονηρός as the technical pair for the second God. And the lexical range of πονηρός runs through troublesome, burdensome, wretched, malicious, without arriving at bad. “Only ‘bad’ does not belong among them.”
Then, on his fourteenth page, he states the position that makes sense of the whole: Marcion “let righteousness stand as the fundamental property” of the Creator while making it, as ius talionis and pure punitive justice, the source of everything bad. From which follows the observation that a Lutheran might have been expected to miss and did not: the “righteous” is the greater enemy, worse than the bad, “because it is aggressive and directly and decidedly keeps men from faith”, what Harnack condenses into the formula lex fidem impedit.
That is not a monotheist protecting his flank. It is a historian reading his sources against his own instinct, and its central argument, which the next section takes up, has never been properly answered.
The Scaffolding Sentence
And yet the same book pulls hard the other way. On his eleventh page Harnack finds in Marcion’s eschatology “the monarchy of the good God that results in the end, despite the dualistic formula.” On his twenty-fourth page he notes with evident approval that Marcion’s most important disciple, Apelles, “found his way back to monotheism, which does after all lie latent in the master’s own teaching.” On the same page he quotes his own confession in Greek: Πιστεύω εἰς θεὸν πατέρα παντοκράτορα, and says he holds it, on religious and historical grounds, against the fundamental idea of Marcionite Christianity. And then, in a footnote on his twenty-fifth page, comes the sentence that gives this section its title:
I therefore see in the personification of the world-creator one of Marcion’s “scaffoldings” which can be dismantled without difficulty.
Scaffolding exists for the sake of the building. Once the building stands, the scaffolding comes down. Harnack is saying that the second Power is a temporary structure around a religious insight, and that the insight survives its removal. Beside the “monarchy … in the end” and the approval of Apelles, a pattern appears. The distinction is preserved historically and dissolved theologically. It is real in the second century and provisional in the truth.
This is where the True Testament parts from him, and the parting is not about tone but about what the good news is.
If the Creator is scaffolding, then the message is finally about grace, and the two Powers were the second-century idiom in which grace happened to be expressed. Remove them and you lose nothing essential. If the Creator is not scaffolding, then removing him removes the news. The Good Father is not the better moral face of the Creator, nor the Creator perfected, nor the deep unity behind two religious experiences. He is another God, not previously known, who has acted in a world that is not his own and taken out people who were not his.
What Two Powers Does Not Mean
I want to be exact about what is and is not being claimed. This is a structural claim about the Evangelion and Apostolikon as the True Testament receives them, not a reconstruction of what the historical Marcion privately thought. And it is not what “two Gods” is usually taken to mean.
It does not mean two equal absolutes. It does not require a symmetrical dualism of matched opposites, light against darkness, in which each Power needs the other to be what it is. It does not mean both are good, and it certainly does not mean both are to be worshipped. What it means is narrower and harder: that the Father of Christ cannot be identified with the maker, ruler and judge of this world.
The asymmetry runs in every direction at once. The Creator owns this aeon and governs it; the Father comes from outside it. The Creator judges by his law, which is his to enforce; the Father rescues, and there is no law in the transaction. The Creator has claims (the world is his work and we are in it) and the Father has none, which is precisely why what he does is grace and not settlement.
The Asymmetry in the Texts
The texts carry that asymmetry in their vocabulary, and the securest of them is the saying about knowing: “Everything has been entrusted to me by the Father. No one knows the Father except the Son, and no one knows the Son except the Father, and anyone to whom the Son chooses to reveal him” (Evangelion 6:30 TT). This saying is well attested as central to Marcion’s case, and its logic is the whole point: the Father is not the God anyone already had. In the preceding verse the True Testament reads “I praise you, Lord of heaven, because what was hidden from the wise and educated you have revealed to the simple” (Evangelion 6:29 TT). The shorter title, without “and of the earth,” is the reading Tertullian and Epiphanius both report, and the True Testament follows them in leaving out “Father” as well. I build nothing more on it here.
The other side of the asymmetry is named directly. “The god of this aeon has blinded the minds of the untrusting, so that the light of the Evangelion of the glory of Christ, who is the image of God, might not shine to them” (2 Corinthians 4:1 TT). The obstacle is not a condition but an agent with an interest in the outcome. And the struggle is placed where an agent of that scale would place it: “For our struggle is not against flesh and blood, but against the powers and the world rulers of this darkness, against the spiritual forces of wickedness in the heavenly places” (Laodiceans 6:5 TT).
Between the two sides stand the transfer texts. Their grammar is the grammar of custody rather than of verdict: the deliverance out of the present evil aeon (Galatians 1:4 TT), and then “He rescued us from the realm of darkness and transferred us into the kingdom of his beloved Son” (Colossians 1:13 TT). And the resulting condition is described as a change of address: “Our citizenship is in the heavens above, from where we eagerly await a rescuer, Christ” (Philippians 3:6 TT). You do not need a second jurisdiction to be forgiven. You need one to be extradited.
One passage has to be handled with more care than it usually gets, because it is regularly conscripted on both sides and it settles less than either party wants. In the True Testament wording, Paul writes: “We know that an idol is nothing in the world, and that there is no God except one. For even if there are so-called gods in the heavens and on the earth, indeed many gods and many lords, yet for us there is one God, the Father, from whom are all things and for whom we are, and one Lord Jesus Christ, through whom are all things and through whom we are” (1 Corinthians 8:1–3 TT).
Read honestly, that does not prove two Powers, and I will not pretend otherwise. It opens by denying that there is any God but one. What it does is something more modest and still worth having. It concedes that others are called gods and lords, in the heavens as well as on the earth. It then identifies, for the assembly, one God, and identifies him as the Father, not as the world’s maker, and one Lord, Christ. The force of the passage lies in that “for us,” and in the fact that the identification is made by relation to Christ rather than by appeal to creation. It will not carry a complete metaphysics, and a confession that leaned on it to do so would deserve the objection it got.
Apelles, and Finding the Way Back
Which brings us to Apelles, and to Harnack’s phrase “found his way back.”
That phrase is an argument disguised as a description. To say Apelles found his way back to monotheism, and that monotheism “does after all lie latent” in Marcion, is to assume the conclusion: that one first principle is where the tradition was always tending, so that a pupil who arrives there has recovered something rather than changed something.
Another possible reading is available, and the earliest witness supports it. Rhodon, a pupil of Tatian writing against the Marcionites, reports that the movement had divided over the number of first principles. Apelles held one. Potitus and Basilicus held two, “as does the mariner Marcion himself.” Syneros led a party holding not two but three natures. On this testimony, which is hostile but early and specific, two principles is the founder’s position and one is his pupil’s departure from it. Rhodon adds that when he pressed Apelles for a demonstration, Apelles said the prophecies refuted themselves, admitted under oath that he “did not understand how there could be a God without being produced, but that he believed it,” and held that one ought not to examine doctrine at all. That is not a man recovering his teacher’s latent meaning. It is a man who has reduced the principles to one while keeping a second source for the prophecies, and who cannot say why.
What is certain is that Marcion’s own church argued the question hard. Christoph Markschies notes that whereas the Valentinians are recorded disputing the theory of first principles in only one place, such disputes among Marcion’s followers are attested considerably more broadly. The debate is often labelled by moderns with the catchword three first principles, though Rhodon’s own phrase for the third position is “three natures.”
That finding cuts two ways, and both are worth having. It supports Harnack’s claim on his nineteenth page that Marcion’s church “never had a common doctrine of first principles, but rather left room here for the most diverse doctrines and teachers”; sustained argument is what an unenforced question produces. But it tells against the stronger version of his claim on the same page, that Marcion “had no interest at all, or only a very slight one, in philosophy-of-religion theories.” The question was live enough in his church to generate more controversy than the Valentinians managed. The absence of an enforced doctrine is not the absence of a pressing question, and Harnack sometimes treats the two as the same thing.
So: formally, Harnack keeps the Powers apart, and he does it better than the critics who accused him of inventing the distinction. Substantively, he does not let the distinction be final. It is historical rather than theological for him, a feature of Marcion’s century rather than a claim about God, and he says as much when he calls the second Power scaffolding.
That leaves the question this section has deliberately not answered. Harnack’s Creator is righteous, punitive, limited, and (in the most striking phrase in the pamphlet) himself harassed. It is a subtle and genuinely argued portrait. But it is also a portrait that keeps the Creator inside the range of the morally intelligible, and Harnack needed him there. A Creator who was simply bad would have made Marcion a mythologist and cost him the argument of the last section. The next question is whether the portrait is drawn from the sources or from that need.