1. To Speak or To Read?

The word ἀλληγορέω occurs fairly late in the Greek corpus, with only a handful of occurrences before the first century ce, most of which are from fragmentary texts. Most scholars interpret it either: (1) to speak allegorically or (2) to read allegorically depending on the context.Footnote ⁴ The difference lies in the agent of the action: in ‘to speak’ the text or the author is allegorising, whereas the reader is the agent of ‘to read’. In definition one, ἀλληγορέω merely denotes a textual trope. In definition two, the interpreter is indicating his method of reading.

2. A Deeper Definitional Problem

The primary evidence contains both of these senses, but it also reveals a level of complexity. For example, if the term means ‘to read allegorically’, then what exactly does ‘reading allegorically’ entail? Matthew Harmon thinks ἀλληγορέω means to read a text in light of some sort of external framework.Footnote ⁵ Steven Di Mattei thinks it means to read something as a this-for-that trope.Footnote ⁶ David Dawson argues that it means to construct an extended narratival metaphor that includes a beginning, middle and end.Footnote ⁷ Although these scholars agree that ἀλληγορέω denotes reading, they do not agree on what sort of reading it entails. The same problem occurs for speaking allegorically. If the word refers to a literary phenomenon, what sort of phenomenon is it? Pierre Bonnard argues that the meaning is closer to typology than Philonic allegory.Footnote ⁸ Gerhard Sellin disagrees, arguing that it means allegory (Allegorie).Footnote ⁹ F. F. Bruce insinuates that the word refers to something like Bunyan's Pilgrim's Progress.Footnote ¹⁰ Both definitions struggle to provide clarity past the basic textual-hermeneutical dichotomy.

Once ἀλληγορέω is understood, one must then decide what modern phrase captures this type of speaking or reading. For example, Steven Di Mattei claims ἀλληγορέω merely refers to a this-for-that trope.Footnote ¹¹ The phrase ‘to speak allegorically’ probably does not capture that meaning because it suggests robust pieces of literature such as Pilgrim's Progress.Footnote ¹² Allegories like Bunyan's classic contain too many essential attributes to be reduced to this-for-that tropes. Therefore, although the basic speak-read dichotomy has survived in scholarship, it leaves many questions unanswered.Footnote ¹³

3. Summary of the Solution

I answer these questions in three ways: First, Paul's use falls within the speaking sense, suggesting the verb refers to something Genesis itself does. The rendering ‘these things are symbolic’ attributes the ‘allegorical’ quality to the text itself. Second, ‘symbolic’ captures how the term ἀλληγορέω functioned in the ancient world. Around the time Paul wrote Galatians, ἀλληγορέω often referred to a broad range of figures of speech such as euphemisms, proper metaphors, synecdoche, similes, parables, and metonyms. The adjective ‘symbolic’ functions similarly in modern English.

Third, ‘symbolic’ preserves Paul's view of the literal sense of Genesis in a way that ‘metaphorical’ does not. Admittedly, the noun ‘metaphor’, used colloquially, maintains the closest sense to ἀλληγορέω.Footnote ¹⁴ It can refer to an entire trope in a way that ‘symbol’ cannot, and it refers to a broad range of textual tropes. The problem with using the adjective ‘metaphorical’ in Galatians 4.24 is the connotation that Paul does not think that the literal sense is true, a problematic connotation ‘symbolic’ does not have. ‘Symbolic’ merely communicates that these narratives have additional significance. Thus, although ‘these things are metaphorical’ comes close to capturing Paul's meaning, ‘these things are symbolic’ shares the same benefits of ‘these things are metaphorical’ because the adjectives share a similar range of meaning, and ‘symbolic’ avoids obfuscating Paul's understanding of Genesis 16–21. Nevertheless, I will use the noun ‘metaphor’ when analysing the use of ἀλληγορέω because of the weakness of ‘symbol’, but I will show why ‘symbolic’ is the best fit for Galatians 4:24.

4. Sense One: ‘To Speak Symbolically’

Up until the end of the second century ce, ἀλληγορέω predominantly meant ‘to speak metaphorically or symbolically’.Footnote ¹⁵ Three pieces of evidence make this clear. First, when the verb means ‘to speak metaphorically or symbolically’, the subject of the active verb is the author or the text itself, not the reader. Second, the examples of ἀλληγορέω would be called metaphors by modern English speakers. Finally, when ancient authors define ἀλληγορέω, they describe a trope that fits the modern notion of metaphor in the broad, colloquial sense.

One of the first occurrences of ἀλληγορέω in the ancient Greek corpus comes from a scholion on Euripides’ Phoenician Women.Footnote ¹⁶ The ancient Greek commentator compares Empedocles, a pre-Socratic philosopher from the middle of the fifth century bce, to Euripides, the famous poet from a few decades later. The scholion reads, ‘“Do not sow the furrow of children”. Empedocles speaks symbolically (ἀλληγορῶν φησι) when he uses the phrase, “the divided meadows of Aphrodite”, by which he means “the genesis of children”. Euripides, speaking of the same thing as Empedocles, flees shameful thoughts, and he uses the metaphors “sowing” and “furrow”’ (Schol. Eur. Phoen. 18. 66.3–5).Footnote ¹⁷ The author is commenting on how both men referred to the female anatomy: Empedocles with ‘the divided meadows of Aphrodite’ and Euripides with ‘the furrow of children’.

In this paragraph, several contextual clues indicate ἀλληγορέω means ‘to speak symbolically’. First, Empedocles is the agent of the action, meaning that ἀλληγορῶν refers to a literary or authorial phenomenon, not a hermeneutical method. Second, the literary technique the commentator discusses is a metaphor. In the author's opinion, Empedocles uses ‘the divided fields of Aphrodite’ as a metaphor for the female anatomy. It is not an allegory but a ‘euphemism’; that is, a metaphor to modestly refer to something considered crude. Third and finally, the parallel drawn between the two authors suggests that ἀλληγορέω refers to the same thing as μɛταφορά (metaphor).Footnote ¹⁸ The author is saying that Euripides is doing the same thing better than Empedocles, thereby binding ἀλληγορῶν φησι (‘he speaks allegorically’) to ἐχρήσατο . . . τɛχνικαῖς ταῖς μɛταφοραῖς (‘he uses skillful metaphors’). Thus, at the time this scholion was written, ἀλληγoρέω meant ‘to speak metaphorically’.Footnote ¹⁹

Although Pausanias was writing in the early part of the second century ce, he was one of the founders of Attic lexicography.Footnote ²⁰ Thus, his discussion of ἀλληγορέω reaches back to the conquests of Alexander the Great. In his work, Ἀττικῶν ὀνομάτων συναγωγή (A Collection of Attic Words), Pausanias defines the term διομήδɛιος ἀνάγκη as, ‘a proverb like that from Tydeus or from Thracian, who compelled the foreigners to sleep with his deformed daughters, whom the sentence represents symbolically (ἀλληγορɛῖ) as horses’ (Pausinias, Ἀττικῶν ὀνομάτων συναγωγή 14.2). Just like above, Pausanias uses ἀλληγορέω to refer to an action of the text. ‘The sentence’ (ὁ λόγος) is the subject of ἀλληγορɛῖ and refers to the episode he just described between Tydeus and the daughters with whom he slept. Thus, the verb likely refers to a literary phenomenon.Footnote ²¹ Second, although it is less clear than the instance from the scholion on Euripides, the idea of ‘metaphor’ fits. The deformed daughters (θυγατέρɛς αἰσχραί) serve as symbols for horses (ἵππους), which modern English speakers would call a metaphor.

Demetrius’ two uses of the term in his book De elocutione are clearer.Footnote ²² In section 151, he discusses how some allegories (ἀλληγορίαι τινές) have a colloquial flavour (τι στωμύλον). In his third example, Demetrius claims that the poet Sophron ‘speaks symbolically (ἀλληγορɛῖ) about women in the following line about fish: “tube fish (σωλήν), sweet-fleshed oysters, dainty meat for widows”’ (Eloc. 151).Footnote ²³ This line from Sophron quoted by Demetrius is a crass joke, playing off the meaning of the word σωλήν or ‘tube fish’ which serves as a double entendre for male anatomy.Footnote ²⁴ This example demonstrates that Demetrius used ἀλληγορέω to refer to a double entendre resembling a modern metaphor.

The second occurrence of ἀλληγορέω in De eloctutione is similar. In section 285, Demetrius quotes the line, ‘a city which is no longer anchored to our ancestors prepared for battle at sea, but an old woman, wearing sandals and gulping down some gruel’ (Eloc. 285). Then Demetrius explains, ‘Here, “old woman” functions as a symbol (ἀλληγοροῦν) for a city that is already weak and fading while at the same indicating its weakness by speaking hyperbolically. Likewise, “gulping down some gruel” functions as a symbol, describing a city that was formerly concerned with meat distribution and banquets while squandering the needs of its soldiers’ (Eloc. 285).Footnote ²⁵ Even without Demetrius’ explanation, modern readers would recognise this as a metaphor.Footnote ²⁶

Trypho most likely wrote in Rome in the second half of the first century bce.Footnote ²⁷ Although he uses the cognate noun ἀλληγορία, his discussion in Πɛρὶ τρόπων provides an ancient definition of ἀλληγορέω and an example.Footnote ²⁸ Trypho claims that ‘an allegory (ἀλληγορία) is a statement that, describing one thing regularly, actually brings to mind the thought of something else according to a likeness with the former thing’. As an example, he quotes ‘whose bronze pours out most straw on the ground’.Footnote ²⁹ Trypho's definition fits the idea of a metaphor. A metaphor is a trope that seems to describe something straightforwardly but actually brings to mind something else by tying the two things together through something they share.

Trypho's example further corroborates this reading. The line he quotes occurs in book 19 of the Iliad in a conversation between Odysseus and Achilles. Incensed by the death of Patroclus, Achilles announces his intent to seek revenge against Hector (Il. 19.199–214). Odysseus intervenes and tries to convince Achilles of the cost of war by arguing that ‘men quickly have their fill of war, whose bronze spills most straw upon the ground’ (Il. 221–2). In other words, men grow tired of war because it is not worth the blood spilled. Odysseus is using straw as a symbol for men of war and bronze as the weapons by which they fall. Trypho claims that this is ἀλληγορία, thus providing evidence that the trope is similar to what modern speakers would call a metaphor.Footnote ³⁰

In, Antiquitates judaicae (e.g., Ant. 1.5), Josephus defends the theological statements in the Pentateuch that would offend Hellenistic Greeks:

Josephus is claiming that Moses may say some confusing things, but he does not describe what such statements might be, nor does he give examples. However, two clues shine light on ἀλληγορέω.

First, the lawgiver is the one who allegorises. The participle ἀλληγοροῦντος occurs within a genitive absolute with τοῦ νομοθέτου (‘the lawgiver’) as its subject. Second, ἀλληγοροῦντος is surrounded by other actions that the lawgiver is performing: Standing parallel αἰνίσσομαι (‘to speak in riddles’), λέγω ἐξ ɛὐθɛίας (‘to speak straightforwardly’), and ἐμφανίζω ῥητῶς (‘to explain literally’). Therefore, ἀλληγορέω must refer to some form of symbolic but understandable speech, although Josephus’ use of the term does not allow an exact understanding of what he meant.

In Philo's work, ἀλληγορέω means to ‘speak symbolically’ multiple times.Footnote ³¹ For example, in De ebrietate on Exodus 32.17, Moses descends Mount Sinai with the tablets of the law. Joshua hears a noise in the Israelite camp and assumes that it is the noise of war (Exod 32.15–17), but Moses tells Joshua that it is singing. Philo, however, defends Joshua: ‘That war was in the camp is very natural, for where else might there be contentions, fights, hostilities and all the works that go with interminable war if not in the life of the body, which speaking symbolically he calls “the camp” (ὃν ἀλληγορῶν καλɛῖ στρατόπɛδον)’ (Philo, Ebr. 99). Before this, Philo discussed the nature of the interaction between the body and the mind. Philo thinks Joshua was right because, since passions are constantly churning in the body, one could properly say that this churning operates as a war. For Philo, Joshua is using ‘the camp’ as a symbol for the human body. The verb ἀλληγορέω, in this instance, does not refer to Philo's reading but to Joshua's symbolic speech.

In De migratione Abrahami 131, Philo ponders the meaning of Deuteronomy 13.4 that commands the Israelites to ‘walk before’ God and decides this cannot be literal because God is not a corporeal being. He concludes that ‘walking before God’ must refer to living in accordance with God's statutes: ‘(Moses) seems to be speaking symbolically (ἔοικɛν ἀλληγορɛῖν), describing a soul's following of God's doctrines’ (Philo, Migr. 131). Here, ἀλληγορέω refers to the action of the author, and that action seems to be what modern English speakers would loosely call a metaphor.

Philo's discussion of the Joseph narrative serves as an example of ἀλληγορέω meaning ‘to speak symbolically’ rather than ‘to speak metaphorically’. After describing Jacob's grief upon hearing the news of his missing son, Philo writes, ‘Surely, it is worth it, after a literal account of the story, to further explain the underlying meaning, for approximately all or most of the law-book is symbolic (ἀλληγορɛῖται)’ (Philo, Ios. 28). Philo's statement suggests that ἀλληγορέω refers to a quality of the text despite being in the medio-passive voice. One should explain the symbolic meaning of the story because the law contains this meaning. The wider context, however, demonstrates that Philo thinks the literal sense is true; therefore, rendering ἀλληγορɛῖται as ‘the law-book is metaphorical’ risks miscommunicating that he thought otherwise.

Writing around the same time as Philo, Heraclitus (not to be confused with the pre-Socratic philosopher) uses ἀλληγορέω twenty-six times, all of which mean ‘to speak allegorically’.Footnote ³² In Allegoriae Homericae, Heraclitus defends Homer against critics who saw him as propagating irreverent myths about the gods.Footnote ³³ Heraclitus repeatedly argues that the poet's critics read him too literally, missing the significance of his metaphors: ‘it is a weighty and damaging charge that heaven brings against Homer for his disrespect of the divine. If he was not speaking symbolically (ɛἰ μηδὲν ἠλληγόρησɛν), he was impious through and through’ (All. 1.1).Footnote ³⁴ Throughout the defence that follows, Heraclitus exhibits all three pieces of evidence. He attributes the action to Homer by making him the subject of the active verb and showing that the metaphors are objectively in the text. Heraclitus’ definition fits with the modern idea of metaphor, and modern English speakers would readily label his examples as metaphors.

Other than the rare occasion when Heraclitus uses ἀλληγορέω as an attributive participle, he uses Homer as the subject of the verb.Footnote ³⁵ Also, throughout his reading of the Iliad and the Odyssey, Heraclitus shows that the metaphors are in the text. For example, Heraclitus asks his interlocutors why Homer might not be allowed the use of such a trope: ‘So, since the trope of allegory (ὁ τῆς ἀλληγορίας τρόπος) is familiar to all and not unknown to Homer, what has possessed us that we shall not mend all the instances in which he seems to think poorly about the gods?’ (All. 6.1).Footnote ³⁶ Clearly, Heraclitus thinks Homer used ἀλληγορία, demonstrating that ἀλληγορία refers to a literary phenomenon.

Heraclitus attempts to provide a cogent definition and multiple examples of ἀλληγορέω, which closely resemble Trypho's: ‘Now perhaps it is necessary for me to provide a concise and careful definition of “allegory” (ἀλληγορίας), for the word itself, having been formed etymologically, implicitly attests to its own meaning. For the trope which speaks (ἀγορɛύων) of one thing but signifies another thing (ἄλλα) other than which it speaks has been named “allegory” (ἀλληγορία)’ (All. 5.1–2). Heraclitus provides multiple examples to illustrate this: Archilochus uses ἀλληγορία when he compares war to a surge of the sea as does Mytilene who compares the woes of tyranny to the sea (All. 5.3–9). Heraclitus even appeals to the same metaphor of straw and copper from Odysseus found in Trypho (All. 5.15). Therefore, Klauck is likely wrong when claiming that ‘in Quintilian, ἀλληγορία designates a literary form; in Pseudo Heraclitus, the process of interpretation steps in’.Footnote ³⁷ Admittedly, the hermeneutic of Heraclitus closely resembles what would come to be known as allegory in the following centuries, but he is adamant that the verb ἀλληγορέω refers to a literary form that closely resembles the modern English notion of metaphor.

This sense continues late into the first century ce and into the second. Plutarch uses the term three times – once to refer to a literary phenomenon and twice to refer to a method of interpretation.Footnote ³⁸ The former occurs in his treatise De esu carnium, in which Plutarch defends vegetarianism. He introduces his opinions by quoting from Empedocles, whose words Plutarch claims served as a veiled critique of carnivorism: ‘Here, (Empedocles) speaks symbolically (ἀλληγορɛῖ) about the soul, namely, that as a judgement for murder, the eating of flesh, and cannibalism, it has been imprisoned in a mortal body’ (De esu 7 (996.B–C)). It is unclear what Plutarch thinks Empedocles is doing, but ἀλληγορέω refers to a mode of composition since Empedocles is the subject of the active verb.

These sources show that ἀλληγορέω quite often meant ‘to speak metaphorically or symbolically’. Each author attributes the action to the author of the text they were reading or to the text itself. The definitions describe the term in a manner that matches the modern idea of metaphor and provide illustrations that are clearly metaphors. This sense seems to have dominated the few centuries surrounding Paul's writing of Galatians 4.24, but there are multiple examples that show that it was not the only sense in which the term was used.

5. Sense Two: ‘To Interpret Allegorically’

In some occurrences, the verb refers to a mode of reading identifiable by the agent of the action. If the reader – as opposed to the text or author – is the agent, then the word does not refer to a textual trope but an interpretive method.

Philo uses the term in this sense multiple times.Footnote ³⁹ For example, in book 3 of Legum allegoriae, Philo pairs ἀλληγορέω with the indefinite pronoun τις, which functions much like the dummy subject ‘one’ in modern English. ‘Let us see next see how someone is said actually to hide himself from God. Were one not to read allegorically (ɛἰ δὲ μὴ ἀλληγορήσɛιέ τις), it would be impossible to accept what has been written down’ (Leg. 3.4). The pronoun τις refers to a reader, meaning that ἀλληγορέω must refer to some form of interpretation. In De migratione Abrahami, Philo claims,Footnote ⁴⁰ ‘Do you not see that the five daughters, whom by allegorising we say are the outward senses (ἃς ἀλληγοροῦντɛς αἰσθήσɛις ɛἶναί φαμɛν), have come from the tribe of Manasseh’ (Migr. 205). ‘We’ serves as the subject of the allegorising, indicating that Philo does not use ἀλληγορέω to refer to a textual trope but to a form of reading.Footnote ⁴¹

Plutarch uses ἀλληγορέω to refer to a mode of reading twice. Both occur in his treatise De Iside et Osiride, in which he attempts to explain some of the more fantastic portions of Egyptian mythology to a priestess named Clea by appealing to parallels within Greek mythology. He claims, ‘Those who think it proper to call the body of the soul “Hades” because it is deranged and drunk inside of it allegorise (ἀλληγοροῦσι) too frivolously’ (Is. Os. 27 (362B)). Leading up to this statement, Plutarch has been discussing various interpretations of Greek mythology. Some argued that the stories originally depicted the kings of old (Is. Os. 22 (359E)), and others thought that they depicted demigods (Is. Os. 25 (360E)). Given this wider context, these frivolous allegorisers are interpreters, not composers, of Greek mythology. Thus, ἀλληγορέω means ‘to interpret allegorically’. The second instance in De Iside et Osiride is similar. After discussing various interpretations of Egyptian mythology, Plutarch suggests looking to writers that think more clearly and philosophically (Is. Os. 32 (363D)). He claims that these sorts of men are ‘like the Greeks who allegorise (ἀλληγοροῦσι) Kronos as time’ (Is. Os. 32 (363D)). Again, Plutarch refers to interpretations of Greek myths, suggesting that the verb is a method of reading.Footnote ⁴²

The only other author in the centuries surrounding Paul that uses ἀλληγορέω to refer to a method of reading is Celsus. His treatise Ἀληθὴς λόγος, probably written in the second half of the first century ce, is one of the earliest known criticisms of Christianity, but only fragmented quotes in Origen's work Contra Celsum have survived.Footnote ⁴³ The instances of ἀλληγορέω attributed to Celsus can be difficult to distinguish from Origen's own voice. For example, in Contra Celsum 1.17, Origen complains that Celsus wrongly faults those who allegorise Moses’ works: ‘In what follows, assailing the history of Moses, (Celsus) finds fault with those who read it figuratively and allegorically (τοὺς τροπολογοῦντας καὶ ἀλληγοροῦντας αὐτην)’ (Cels. 1.17). ἀλληγορέω refers to a method of reading because the history of Moses is the object of the verb, but it is unclear if these words reflect the voice of Celsus.Footnote ⁴⁴ They read more like a paraphrase of Celsus, meaning the use of ἀλληγορέω would reflect Origen's later use not Celsus’ in the second century ce.

Book 4, however, seems to provide a direct quote Celsus that includes ἀλληγορέω. Origen writes, ‘As if solely to hate and despise, (Celsus) has devoted himself to the statements of the Jews and the Christians, saying that “the most reasonable of the Jews and the Christians, being ashamed of such things, attempt somehow to allegorise them (πɛιρῶνταί πως ἀλληγορɛῖν αὐτά), but these things are not to be taken as some allegory but are straightforwardly mythological”’ (Cels. 4.48).Footnote ⁴⁵ If truly from Celsus, this quote would provide evidence that ἀλληγορέω referred to a method of reading in the middle of the second century ce. Although it is not obvious from the quote, the pronoun αὐτά, which is the object of the verb ἀλληγορɛῖν, probably refers to Christian and Jewish writings. Celsus would then be claiming that Christians and Jews used a certain method of reading, ἀλληγορέω, to avoid the uncomfortable parts of their Bibles. These examples from Philo, Plutarch and Celsus demonstrate that ἀλληγορέω did not exclusively mean ‘to speak symbolically’ but, on multiple occasions, referred to a method of reading.

6. The Meaning of ἅτινά ἐστιν ἀλληγορούμɛνα

Based on the texts above, the basic dichotomy between speaking and reading in the scholarly discussion of Galatians 4.24 fits the data. The question, though, is which of these did Paul mean?

7. Reading or Speaking?

The reading view accounts for the medio-passive form of the verb. A. T. Hanson argues that the voice of ἀλληγορούμɛνα indicates a shift from speaking to reading, a phenomenon not uncommon in ancient Greek.Footnote ⁴⁶ However, more compelling evidence commends the speaking view. First, the sense ‘to interpret allegorically’ is relatively rare and late.Footnote ⁴⁷ Philo only uses the term in this sense about nine times, Plutarch twice, and Celsus once, and the latter two authors were writing almost a century after Paul wrote Galatians. Thus, although context ultimately determines the meaning of a word, contemporary Greek literature slightly favours the speaking view.

Second, the medio-passive form of ἅτινά ἐστιν ἀλληγορούμɛνα is not sufficient evidence because the form could be truly passive (‘these things were written symbolically’) or stative (‘these things are symbolic’). The stative rendering would emphasise the ‘symbolic’ character of the story, like Philo's description of the Joseph narratives. Philo's point is to ground his symbolic interpretation in the fact that the law-book has a symbolic quality to it, and the medio-passive form ἀλληγορɛῖται helps communicate this point. Paul's use of the term in Galatians 4.24 occurs in a similar context. He claims that by reading Genesis too literally the Judaizers failed to understand the true significance of the narratives.

The passive reading would have a similar effect. The sentence ‘these things were written symbolically’ could place emphasis on the subject. As Porter argues, ‘the frequent result of the use of the passive voice is that attention regarding the action is placed upon the grammatical subject (recipient) rather than the agent’.Footnote ⁴⁸ However, context suggests that the emphasis would fall on the symbolic quality of the text much like the stative rendering. Both Paul and his opponents are reading the same Abraham narratives, making the identity of the grammatical subject irrelevant to his point. Paul's issue is that his opponents are missing the narrative's significance (cf. Gal 4.21). The truly passive reading, therefore, would read ‘these things were written symbolically’, not ‘these things were written symbolically’. Since the medio-passive form of the verb could support any of these views, the voice of ἅτινά ἐστιν ἀλληγορούμɛνα seems inconsequential in determining the sense of ἀλληγορέω.

Most important in determining the sense are Paul's opening words in Gal 4.21, ‘tell me, you who want to be under law, do you not hear the Law?’ If Paul beckons his interlocutors to hear the text, he must think the text is speaking. The text speaking in verse 21, pushes the phrase in the direction of ‘to speak symbolically’.Footnote ⁴⁹ Thus, the text serves as the agent of the action, and ἀλληγορέω does not refer to Paul's method of reading.

8. Allegorical, Metaphorical, or Symbolic?

The last question that must be answered concerns the nature of the action. Should one render ἀλληγορούμɛνα as ‘allegorical’, ‘metaphorical’, or ‘symbolic’? The modern word ‘allegory’ does not fit the examples from the ancient world. The verb ἀλληγορέω referred to simple metaphors, euphemisms, double entendre, and other figures of speech. It did not refer to robust metaphorical narratives.

Used colloquially, ‘metaphorical’ refers to a similar range of tropes, effectively meaning little more than ‘non-literal’ in modern English. Unfortunately, this rendering risks communicating that Paul thought the Abraham narratives were not historically true, which is not the case. Paul begins the pericope by paraphrasing the events of Genesis 16–21 as if they were true (Gal 4.22–23), and his argument in Galatians 3.17 hangs on the historical sequence of Abraham's promise and Mosaic law. The law does not nullify the promise because it came 430 years after. Paul did not think the Abraham narratives were mere metaphors. The adjective ‘symbolic’, therefore, fits best because it shares the broad sense of ἀλληγορέω and avoids the problems caused by the connotations associated with ‘metaphorical’.

Paul uses the phrase to call out his opponents for missing the significance of the story of Ishmael and Isaac. The circumcision party presumably read Genesis 16–21 like Jubilees 16.17: ‘All the seed of [Abraham's other] sons should be Gentiles, and be reckoned with the Gentiles; but from the sons of Isaac one should become a holy seed, and should not be reckoned among the Gentiles’.Footnote ⁵⁰ In other words, true sons of Abraham consisted of Isaac's circumcised descendants (Gal 5.2, 6.13). Paul uses ἅτινά ἐστιν ἀλληγορούμɛνα to accuse these opponents of reading Genesis 16–21 too woodenly. Yes, Isaac was Abraham's true son, and modern-day Jews were his progeny. However, Abraham had two sons, not just one (Gal 4.21), and they were both circumcised. If circumcision were what made one a true child of Abraham, how can Ishmael be excluded or reckoned among the Gentiles? The story of Ishmael and Isaac contains significance that these Judaizers missed. The phrase ἅτινά ἐστιν ἀλληγορούμɛνα draws attention to this fact, whether stative or passive.

9. Implications

At least since Origen's day, scholars have attempted to categorise Paul's hermeneutic in Galatians 4.21–31. Was he employing a method of interpretation developed by the Greeks that has now come to be known as ‘allegory’? Understanding the precise meaning of ἅτινά ἐστιν ἀλληγορούμɛνα is crucial to answering this question. The affirmative answer emphasises Paul's use of ἀλληγορέω, arguing that the apostle was intentionally flagging his reading as allegorical. The negative answer often explains the term away. Steven Di Mattei, for example, claims that Paul's hermeneutic is the same allegory represented by authors like Philo, and he grounds this in the meaning of ἀλληγορέω.Footnote ⁵¹ Interestingly, Di Mattei's reading of the phrase is similar to my own, agreeing that the word refers to a textual phenomenon, but if this is true, then Di Mattei's conclusion does not seem to follow.Footnote ⁵² Paul's claim would be an innocuous statement about the text that would not be sufficient to bridge that gap between him and an allegorical hermeneutic like that found in Philo.

On other hand, those who do not think Paul was using an allegorical hermeneutic struggle to provide a meaning for ἀλληγορέω that fits its wider use. In his classic comment on Galatians 4.24, Chrysostom claims that Paul misspoke. ‘With a misuse of language (καταχρηστικῶς), Paul calls a type an allegory’.Footnote ⁵³ Iain Provan defends a modern version of Chrysostom's view, but argues that Paul was not using ἀλληγορέω as it was typically used in the ancient world. In his words, ἀλληγορέω ‘does not signify in Paul what it signifies in other ancient authors like Philo and his later fellow Alexandrians (such as Origen). It signifies rather the kind of typological/figurative reading that . . . is best considered as one aspect of what is involved in the literal (and canonical) reading of Scripture’.Footnote ⁵⁴ Both authors emphasise the context of the statement, arguing that Paul's use of the term is unique. Although context plays a significant role in determining the meaning of any word, it would be odd for Paul to use a common verb in a way that was completely novel. My thesis, however, allows for the possibility of readings like Chrysostom's and Provan's that recognise the differences between Paul and Philo's hermeneutics, but also does justice to the wider use of the term. Paul may be reading the text typologically as opposed to allegorically, but ἀλληγορέω does not refer to this method of reading. In fact, it does not refer to a method of reading at all. Paul is calling attention to the fact that his opponents were missing the significance of the story. He is not flagging his allegorical hermeneutic, but neither is he flagging a typological hermeneutic. Paul's hermeneutic must be determined by what he does with the text, not his use of ἀλληγορέω.