G. Anthony Keddie
In an impassioned editorial in The Guardian, the journalist and scholar Moustafa Bayoumi proclaimed, “I’m a brown Arab-American and the US census refuses to recognize me” (Bayoumi, 2019). The 2010 US census categories, which are being reused in the 2020 census despite considerable dissent, separate ethnicity and race and ask only about Hispanic, Latino, or Spanish ethnicity. The census questionnaire allows for four different races—White, Black or African American, American Indian or Alaskan Native, and “some other race.” Records from the 2000 and 2010 censuses show that the “others” in this fourth category constitute the third largest racial demographic in the United States, and that those who selected this category were, in fact, 97 percent Hispanic, Latinx, or Spanish. Bayoumi noted that the race categories on the census not only force Latinxs into a category of “others,” but they also misclassify other racial groups. In his words,
According to the Office of Management and Budget’s classifications, “white” refers to a person having origins in any of the original peoples of Europe, the Middle East, or North Africa. My origins are from Egypt. Every day, I live my life in America as a brown person. Defining me as white and likening me to a European, as the census does, is absurd.
(Bayoumi, 2019)
In the months following the publication of Bayoumi’s editorial, a federal judge prevented the Trump administration from adding a citizenship question to the 2020 census in what the leaked Hofeller files have exposed as a nefarious attempt to aid Republican redistricting efforts by scaring immigrants away from voting (Wines, 2019). As a result, it is clearer than ever that censuses are not an innocuous tool for collecting population statistics, for counting people.
Following recent debates about the US census while researching the economics of the Roman census has been unsettling, yet revealing. I now see that my positions of privilege as a white, male, US citizen have enabled me to overlook many of the ways that censuses institutionalize discriminatory discourses pertaining to gender, class, status, ethnicity, race, and citizenship. Censuses have been used in the United States to block people of racially mixed ancestry from marrying whites (the so-called “one-drop rule”), to determine who counted as “Jewish” in Nazi Germany, to identify who counted as “Tutsi” during the genocide in Rwanda, and so on.2
Despite the widespread use of modern censuses to marginalize certain groups, scholars have scarcely explored this as a potential dimension of the Roman census. Mainstream New Testament scholarship has been too preoccupied with determining the authenticity of Luke’s narrative about the Augustan census.3 Beyond this search for authenticity, only two consequences of the census are usually taken into account: first, that Roman censuses were symptomatic of an increased tax burden under Roman rule; and second, that the Roman census symbolized the loss of national sovereignty through subjection to direct Roman rule (among others: Dunn, 2003: 310; Horsley, 2006). In this regard, many scholars have interpreted the impact of the census through the lens of Josephus’s account of the Zealots emerging in response to the 6 CE census. If we only had Josephus, it would indeed appear that the census caused increased taxation and provoked widespread resentment toward Roman rule that culminated in the First Jewish Revolt.4
Fortunately, however, Egyptian and Arabian census documents can help us to fill out this picture. These are the main sources that I investigate in this paper. I begin by examining the economic implications of provincial censuses in the Roman East before turning to their sociopolitical effects. Inspired by Bayoumi’s call to recognize how the US census regulates racial identities, I then proceed to identify ways in which the New Testament passages on the census may reflect this institution’s marginalization of women, Jews, and other disadvantaged persons.
There are substantial gaps in our knowledge of Roman censuses, but one thing is very clear: census practices differed significantly from province to province and even from village to village. Luke’s description of a universal Augustan census advances his own perception of how the census was organized. Documentary and inscriptional evidence shows, however, that censuses were administered differently by each province (Le Teuff, 2012: 299–344).5 Augustus extended the census throughout the empire, but his provincial governors and their subordinates were responsible for conducting each province’s census.
In the imperial era, the earliest known provincial census was taken in Gaul (and probably Spain) in 27 to 25 BCE (following a census of just Roman citizens in 28 BCE) (Le Teuff, 2012: 15–6). Some scholars have supposed that the Egyptian census, known as the “house-by house registration” (kat’ oikian apographē), began around the same time, but 11/10 BCE is the earliest Egyptian census date known from the papyri. Securely dated documents indicate that the census was taken every seven years before a fourteen-year cycle began in 19/20 CE (Claytor and Bagnall, 2015). The eventual standardization of a fourteen-year cycle is an important clue that one of the main functions of the Egyptian census was to collect poll taxes: men were required to pay the poll tax from the age of 14, so the fourteen-year census ensured that all minors who would become eligible for the poll tax during the coming cycle would be counted.
Recent research by Andrew Monson (2012; 2014a; 2014b), among others, has problematized two traditional scholarly assumptions about the Egyptian census: first, that it was a Roman innovation; and second, that it was used as a tool of economic exploitation in the form of increasing the tax burden. In the first century BCE, Ptolemaic officials used censuses to collect a range of capitation taxes, such as the syntaxis and epistatikon. Much like under Augustus, Ptolemaic censuses and their related poll taxes were sometimes given the name laographia. A petition from an Egyptian priest to the prefect in 5/4 BCE assumes outright continuity between a type of laographia under Cleopatra and the one taken under Roman rule:
To Gaius Turranius from Soterichos son of Nouchis, a priest chosen by his fellow priests Haryothes, Harsiesis, and Soterichos, priests of Isis, Sarapis, Harpsenesis, and Asklepios in Bousiris of the Herakleopolite nome beyond Memphis. Great governor, formerly even in the time of the queen [Cleopatra VII] until the 25th year of divine Caesar Augustus [6/5 BCE], we four priests out of the six who perform sacred services and carry out purifications and sacrifices two by two were undisturbed on account of the poll tax [laographia] but in the present 26th year of Caesar [5/4 BCE] because Epiodoros the royal scribe of the nome registered us for collection we were pressed into payment on account of the poll tax [laographia] for four years, each of us 64 silver drachmas [16 each], and on account of dike-taxes 108 silver drachmas. We therefore ask…
(BGU IV 1198 [5/4 BCE, Herakleopolite nome]; trans. Monson, 2014a: 150)
Acting as a representative for his fellow priests, Soterichos pursued the continuation of an exemption from the laographia for priests that was in place under Cleopatra VII, prior to Roman annexation. The request assumes that this exemption was maintained as part of the local census practice after annexation but was ignored for the first time in 5/4 BCE.
Literary and documentary sources show that the Ptolemaic censuses and concomitant capitation taxes had been highly irregular and were often abused by officials (Keddie, 2016: 204–6). Where sufficient data are available, it is now clear that the Augustan poll tax (most often 16 drachmas) represented either the same or a lesser burden than the Ptolemaic head taxes. In the Herakleopolite nome, for instance, it appears that the Roman government reduced the late Ptolemaic rate of 20 drachmas or higher per year to 16 drachmas (Monson, 2014a). These Egyptian data should make us cautious about assuming that any provincial censuses increased the local tax burden. Relative continuity was more likely the norm.
The declarations from the 127 CE census in Arabia suggest that the main purpose of censuses in this province was to collect land tributes, not poll taxes as in Egypt. The following declarations employ much of the same formulaic language, though gender may account for some key differences.6
I, Sammouos son of Shim’on, of Maoza in the Zoarene district of the Petra administrative region, domiciled in my own private property [oikōn [e]n idiois] in the said Maoza, register myself [apographomai emauton], thirty years old, [as owner of a] half-share of a field…
(P.Ḥev./Se. 62 [127 CE]; trans. Cotton, from Cotton and Yardeni, 1997, modified)
I, Babatha daughter of Simon, of Maoza in the Zoarene district of the Petra administrative region, domiciled in my own private property [oikousa en idiois] in the said Maoza, register what I possess [apographomai ha kektēmai] (present with me as my guardian being Judanes son of Elazar, of the village of En-Gedi in the district of Jericho in Judea, domiciled in his own private property [oikountos en idiois] in the said Maoza).
(P.Yad. 16 [127 CE]; trans. Lewis, 1989, modified)
On the basis of these two declarations, there is a debate about whether Babatha, and thus women in Arabia, were required to pay a poll tax. According to the third-century CE jurist Ulpian (Dig. 50.15.3), women paid poll taxes in Syria. This has led Udoh (2005: 218, 229), Cotton (2003), and others to conclude that they likely paid poll taxes in second-century CE Arabia (unlike in Egypt at the same time). The different language employed in the highly formulaic declarations of Babatha and Sammouos is striking, however. Whereas Sammouos’s declaration says “I … register myself, thirty years old…,” Babatha’s says “I … register what I possess,” names her legal guardian, and does not list her age. I am not convinced that we have enough information to say with any certainty whether women paid the poll tax in second-century CE Arabia, but the evidence that we do have suggests that they might not have done so (Lo Cascio, 1999: 201; cf. Lewis, 1988: 136; Cotton, 2003: 114–5; Udoh, 2005: 218; Le Teuff, 2012: 160–62).
Apart from the question of gender, the fact that these two declarants mentioned where they were domiciled and, at least in the case of men, their age, shows that the Arabian registrations involved more than just land tributes paid in kind. The demographic information these registrations collected would also have been used to continue the collection of certain Nabataean monetary taxes.7 Unfortunately, based on the extant sources, we have no way of knowing how the rate of the land tribute in Arabia compared to its Nabataean antecedent.
The Syrian governor (technically, legatus Augusti pro praetore) Quirinius conducted the first Roman census of Judea in 6 CE. It involved the registration of property and thus served the purpose of collecting land tributes. Like its Arabian counterpart, this census collected information beyond what was necessary for land taxes. Note, for instance, that the epitaph of Quintus Aemilius Secundus recognizes an official’s part in counting 117,000 people in Apamaea during Quirinius’s census of Syria, presumably part of the same 6 CE census that impacted Judea: “On the orders of Quirinius, I was also the person who conducted the census of the city of Apamaea: 117 thousand citizens of both sexes” (iussu Quirini censum egi Apamenae civitatis millium homin(um) civium CXVII) (trans. Hall, 2004: 49, modified).8 How would this population statistic have been known if the census were only for property registration? We cannot be certain that the Judean censuses were also used for capitation taxes, but I suspect that the information they collected was indeed used to resume the collection of Herodian capitation taxes; for example, the “house tax” remitted by Herod Agrippa I according to Josephus (A.J. 19.299).
What does this lacunose record of provincial censuses tell us about their functions and consequences? If we consider the Roman provincial census only as a mechanism for collecting taxes, we should conclude that it improved economic conditions in the provinces: it made tax collection more efficient, placed checks on administrative corruption, and standardized taxes at a rate equal or lower than that of previous regimes. In this sense, the provincial census was an institution whose function was economic regulation.
Scholarly theories of regulation are a nexus at which there is considerable convergence between New Institutional Economics and the Marxist Régulation School (Labrousse and Weisz, 2013). Both approaches are helpful for interpreting the institutional impact of the census. Proponents of the former identify the ways that institutions gradually reduce economic uncertainty in society by structuring human interaction. They argue that informal institutions such as cultural norms induce socioeconomic continuity and stability, even after political upheavals (among others: North, 1990). This New Institutional framework makes sense of the continuity between Ptolemaic and Roman census practices and the potentially positive effects of the introduction of regular census intervals and poll tax rates. Yet, the dialectical framework of Marxist régulation theory rightly reminds us that regulation is not always a blessing. As Roland Boer and Christina Petterson observe, “The solution is at one and the same time the cause of further problems” (2017: 46).
If the purpose of the census were only to collect head and land tributes, there would be no reason for census declarations to list people who were not liable to these taxes. In Egypt, only men between the ages of 14 and 62 had to pay the poll tax, yet census declarations also included women, minors, and elders. This is because the census was also a broad-ranging institution of population control, which generated and maintained norms. Building on Foucault’s (2007; 2008) interpretation of the census as a modality of biopower aimed at population control, some scholars of racial politics have described the US census as institutionalized discourse that prejudicially distributes privileges, regulates bodies, and shapes subjectivities (Nobles, 2000; Sandset, 2018). The census does not reflect reality; it constructs it. In different ways, this was also true of Roman provincial censuses.
The Egyptian census declarations list age, filiation, distinguishing physical marks such as scars and even moles, certain types of legal status, and in some cases professions (Capponi, 2005: 92). One of the most striking features of the census declarations is that they were often filed by the male head of the household (pater familias) on behalf of those in his home, his social subordinates. When there was no male head of the household, a female would register under the aegis of a legal guardian, as in the land registration of Babatha presented above and, for instance, in the census declaration of Tatubynchis of Philadelphia:
To Eirenaios and Maron and Herakleides and Ammonios and Petesouchos, laographoi, and to Herakleides, village clerk of Philadelphia, from Tatybynchis, daughter of Mareis, of the same village, with my guardian, my relative Patouamtis, son of Ptollis. In the present twentieth year of Tiberius Caesar Augustus I register my son Panetbeueis, son of Kephalon, who lives in my own house; my above-mentioned son Panetbeueis, son of Kephalon, is 5 years old, and myself, Tatybynchis, daughter of Mareis, 35 years old, with a scar on my right thumb, together with my guardian, my relative Patouamtis, son of Ptollis, who is 36 years old and has a scar below his right brow.
(SB I 5661 [34 CE, Philadelphia]; Huebner, 2019: 39)
After census declarations were received, tax officials glued them together into tax rolls called tomoi (Bagnall and Frier, 1994: 19–21, 28). This demographic information then served as a basis for generating lists of people liable to certain taxes or public liturgies and lists of those eligible for tax discounts. It also served as an official record for establishing inheritance rights.9
The only surviving document that might be related to the Judean census appears to be some sort of census-derived register. Discovered at Naḥal Ṣe’elim, this document was likely composed in Judea in the first half of the second century CE (though an Arabian origin is also possible). It lists at least 55 men along with their ages, all between the ages of 13 and 67 (see Table 4.1).
Col. i |
Col. ii |
Col. iii |
---|---|---|
trace son of [ o]s son of [Shim‘]on (?) |
[age] 25 [ag]e 13 [ag]e 19 [ag]e 41 [age] [age] [age] age 61 age 67 [ ] [ ] [age] 22 [age] 36 [age] [?]2 |
[….]..[ Yeshu‘a, son of Levi […]nor [son] Yosepos [another son] Yeshu‘a, an[other son] Yosepos [another son] Aneinas [another son] Ellelos, an[other son] Gaius, a[nother son] Se..os, son of Seima[ A[ K[ |
Hannah Cotton has demonstrated, through comparison with Egyptian tax registers, that the names are arranged by household (see Cotton’s critical edition of the document in Charlesworth, 2000; cf. Cotton, 2003). The head of the household is listed first with a patronymic and is followed by the names of other male members of the household, which are indented by one centimeter.11 Whether this very fragmentary document was indeed a tax register or served a different purpose (e.g., registration for public liturgies), the information on it was likely derived from census declarations.
In Egypt, laographia declarations appeared very similar to epikrisis declarations, and these terms were sometimes used interchangeably in official documents (Wallace, 1938: 109–12; Capponi, 2005: 92–5). Technically, an epikrisis was an examination of special statuses in order to determine tax discounts. Most of the same officials who oversaw the census were also involved in the epikrisis.
The epikrisis was an extension of the census that discriminated according to ethnicity. As Broux (2013; 2015) and others have argued, the Roman epikrisis practices gradually transformed the Ptolemaic head tax privileges for “settlers” (katoikoi) and those called “Greeks” (hellēnes), who may not have self-identified as Greek, into privileged rates for urban elites categorized as “metropolitans” or part of the gymnasial order.12 Consider the following first-century CE record of an epikrisis of metropolitans and the gymnasial order in Hermopolis Magna, for example:
To Antonius Ptolemaios, strategos of the Hermopolite nome. Those minors of the metropolis entering the age-group of fourteen years old in the oktadrachmoi [i.e., an order liable to pay only half of the poll tax rate—8 drachmas] and those of the order of the gymnasium, since they have to be scrutinized [i.e., epikrisis], if they preserve their metropolite descent from both parents, those from the gymnasium, if they are from that order itself. For the epikrisis of these (boys) it will be most necessary that it happens by worthy men of [70 (?)] years old.
(SB V 8038 [first cent. CE, Hermopolis Magna]; trans. Broux, 2013: 147–8)
The institutional discourse advanced by provincial census and epikrisis practices thus construed Roman citizens and Alexandrian citizens, who were tax exempt, as superior to metropolitans and the gymnasial order in other Egyptian cities, who received tax discounts, and both of these brackets as superior to the majority of the population of Egypt (Broux, 2013).13
An epikrisis was not just a mechanism for examining privileged statuses, however. According to the lengthy Arsinoe papyrus of 73 CE, it was also used to collect the Ioudaiōn telesma, the tax exacted from Ioudaioi (“Judeans”/“Jews”) throughout the empire following the First Revolt in Judea (Wallace 1938, 170–6; Heemstra, 2010; Keddie, 2018):
The report of Herakleides, amphodarchēs of the quarter of “Apollonios Camp.” The liability for the Jewish tax for the fifth year of the Emperor Caesar Vespasian Augustus being an abstract according to the statement of the fourth year. The number of Jews taken up by previous accounts are 5 adult males, 6 adult females, of whom one is over age and was adjudged as such, being 59 years of age in the fourth year, one minor, being four years of age in the fourth year, total of names 12. And those taken up through a transcript of the preceding epikrisis shown to be three years of age in the fourth year, being one year old in the second year. Males: Philiskos, son of Ptollas, grandson of Philiskos, mother Erotion. Females: Protous, daughter of Simon, son of Ptolemaios, mother Disarion, total 2, 14 in all, of these adult males 5, minor male 1, who in the fifth year was five years old, likewise 1 minor female four years old … total of names 14. In addition there is enrolled in the Jewish tax in the fifth year of Emperor Caesar Vespasian Augustus, of minors who in the third year were one year old and so in the fifth year three years old, of males taken up, Seuthes, son of Theodoros, grandson of Ptolemaios, mother Philous found to be two years old in the fourth year, in the epikrisis of the fourth year; total of names 15; of which adult males 5, 1 minor male who in the fifth year is five years old, adult females 6, 1 female minor who in the fifth year is five years old, adult females 6, 1 female minor who in the fifth year is five years old, likewise 1 four years old, total 15. Of these 5 adult males being of those in the list as liable to the maximum rate of the laographia, 5 names, and the remaining names 10….
Together with the 5 names of those liable to the maximum rate of the laographia, total as above 15 names, at 8 drachmai and 2 oboloi each, 125 drachmai, for the aparchai 15 drachmai, total 140 drachmai. A similar copy has been deposited with the royal scribe through Amoutio … the scribe.
(CPJ 2:421 [73 CE, Arsinoe]; trans. Tcherikover, 1957)
Whereas Ioudaioi were not required to list their ethnicity on census declarations, and they were issued special receipts for paying the Jewish tax, the Arsinoe papyrus shows that laographia and epikrisis data were used to exact this discriminatory tax. The amphodarchēs of Arsinoe who issued this report was responsible for creating lists of those eligible for all taxes, including the Jewish tax. Notably, this official indicates that he referred to “previous accounts” and “the preceding epikrisis” to determine the ages of those liable to a Jewish tax on top of the poll tax. Indeed, it appears from this document that local tax administrators closely examined information recorded on the laographia rolls to determine liability for the Jewish tax. This is noteworthy because the Jewish tax was collected from both men and women starting at age 3. Here, then, we can glimpse the dark side of regulation: by collecting census information that did not pertain to poll tax liability, the Roman administration was well equipped to exact a punitive tax on Jews, including women and minors. The imperial administration devised this tax to target Ioudaioi throughout the empire (regardless of whether they ever set foot in Judea) as culpable for the First Revolt. The tax not only cast all Jews as outsiders and potential rebels, but it also required them to show allegiance to the empire and its gods in the form of a tax to Jupiter Capitolinus, a god whom they rejected.14
Another corpus in which we can detect the repressive potential of the census institution is the Decian libelli of 250 CE. In response to Decius’s edict that all citizens (except possibly Jews)15 must offer a sacrifice, pour a libation, and partake of the offerings, the pater familias of a household wrote a petition to local officials to certify their declaration with a countersignature. The documents share much in common with the form of a census declaration, probably because they relied on the same bureaucratic apparatus maintained for the census: first, the pater familias issues the document on behalf of the entire household; second, the documents use stock phrases from the census declarations, such as “in accordance with what has been ordered” (and one of the documents is even labeled apographē on its verso);16 third, the performance of allegiance to the emperor through sacrifice is an expansion—albeit a substantial one—of the oath sworn by the divine emperor in some of the census declarations (Rives, 1999: 148–50; Luijendijk, 2008: 167; Blumell and Wayment, 2015: 376). For example, an Egyptian census declaration from 12 CE records the following oath: “I, Harthotes, swear verily by Caesar Imperator Eleutherios, son of god, Zeus Augustus that I have submitted the previous declaration salutarily and truthfully, reserving nothing; if I have sworn truthfully may it be well with me, if falsely, the opposite” (P.Mil. I 3 + P.Col. inv. 8 [12 CE, Theadelphia]; trans. Bagnall, 1991). Notably, even the Jewish woman Babatha had to make such an oath “by the genius of the lord Caesar” when registering her land in Maoza. The surviving Greek translation of her Aramaic oath reads, “I, Babatha daughter of Simon, swear by the genius of our lord Caesar that I have in good faith [omnymi tychēn kyriou Kaisaros kalēi pistei] registered as has been written above…” (P.Yad. 16 [127 CE]; trans. Lewis, 1989). Taking such an oath, though formulaic, would have presented many Jews and Christians with a moral dilemma, as we know the later Decian libelli did for Christians.
Far beyond its regulation of earlier forms of registration and taxation, then, the provincial census was a powerful institution that relied on and reproduced a range of social distinctions, normalizing both privilege and marginalization.
Discussions of the handful of references to taxes and censuses in the New Testament tend to focus on critical issues pertaining to negotiations of imperial power and degrees of economic exploitation. In this section, I raise questions that begin to take account of the sociopolitical dimensions of censuses, complementing and moving beyond these issues. This preliminary survey focuses on passages in the Synoptic Gospels and the Acts of the Apostles that explicitly employ terminology invoking the census, but subsequent studies will be needed to expand this type of investigation with respect both to these passages and to other New Testament texts (e.g., Rom 13:6–7).
Jesus’s teaching on taxes is a useful place to begin this investigation since it appears in each of the Synoptic Gospels. Because their subtle differences in language prove significant, Table 4.2 presents the full Greek text and English translation of each in parallel columns for comparison.17
Mark 12:13–17 |
Matt 22:15–22 |
Luke 20:19–26 |
---|---|---|
13 Then they sent to him some Pharisees and some Herodians to trap him in what he said. 14 And they came and said to him, “Teacher, we know that you are sincere, and show deference to no one; for you do not regard people with partiality, but teach the way of God in accordance with truth. Is it lawful to give kēnsos to Caesar, or not? 15 Should we pay them, or should we not?” But knowing their hypocrisy, he said to them, “Why are you putting me to the test? Bring me a denarius and let me see it.” 16 And they brought one. Then he said to them, “Whose head is this, and whose title?” They answered, “Caesar’s.” 17 Jesus said to them, “Return to Caesar the things that are Caesar’s, and to God the things that are God’s.” And they were utterly amazed at him. |
15 Then the Pharisees went and plotted to entrap him in what he said. 16 So they sent their disciples to him, along with the Herodians, saying, “Teacher, we know that you are sincere, and teach the way of God in accordance with truth, and show deference to no one; for you do not regard people with partiality. 17 Tell us, then, what you think. Is it lawful to give kēnsos to Caesar, or not?” 18 But Jesus, aware of their malice, said, “Why are you putting me to the test, you hypocrites? 19 Show me the coin of the kēnsos.” And they brought him a denarius. 20 Then he said to them, “Whose head is this, and whose title?” 21 They answered, “Caesar’s.” Then he said to them, “Return therefore to Caesar the things that are Caesar’s, and to God the things that are God’s.” 22 When they heard this, they were amazed; and they left him and went away. |
19 When the scribes and chief priests realized that he had told this parable against them, they wanted to lay hands on him at that very hour, but they feared the people. 20 So they watched him and sent spies who pretended to be honest, in order to trap him by what he said, so as to hand him over to the jurisdiction and authority of the governor. 21 So they asked him, “Teacher, we know that you are right in what you say and teach, and you show deference to no one, but teach the way of God in accordance with truth. 22 Is it lawful for us to give phoroi to Caesar, or not?” 23 But he perceived their craftiness and said to them, 24 “Show me a denarius. Whose head and whose title does it bear?” They said, “Caesar’s.” 25 He said to them, “Then return to Caesar the things that are Caesar’s, and to God the things that are God’s.” 26 And they were not able in the presence of the people to trap him by what he said; and being amazed by his answer, they became silent. |
13 Καὶ ἀποστέλλουσιν πρὸς αὐτόν τινας τῶν Φαρισαίων καὶ τῶν Ἡρῳδιανῶν ἵνα αὐτὸν ἀγρεύσωσιν λόγῳ. 14 καὶ ἐλθόντες λέγουσιν αὐτῷ· διδάσκαλε, οἴδαμεν ὅτι ἀληθὴς εἶ καὶ οὐ μέλει σοι περὶ οὐδενός· οὐ γὰρ βλέπεις εἰς πρόσωπον ἀνθρώπων, ἀλλ᾽ ἐπ᾽ ἀληθείας τὴν ὁδὸν τοῦ θεοῦ διδάσκεις· ἔξεστιν δοῦναι κῆνσον Καίσαρι ἢ οὔ; δῶμεν ἢ μὴ δῶμεν; 15 Ὁ δὲ εἰδὼς αὐτῶν τὴν ὑπόκρισιν εἶπεν αὐτοῖς· τί με πειράζετε; φέρετέ μοι δηνάριον ἵνα ἴδω. 16 οἱ δὲ ἤνεγκαν. καὶ λέγει αὐτοῖς· τίνος ἡ εἰκὼν αὕτη καὶ ἡ ἐπιγραφή; οἱ δὲ εἶπαν αὐτῷ· Καίσαρος. 17 ὁ δὲ Ἰησοῦς εἶπεν αὐτοῖς· τὰ Καίσαρος ἀπόδοτε Καίσαρι καὶ τὰ τοῦ θεοῦ τῷ θεῷ. καὶ ἐξεθαύμαζον ἐπ᾽ αὐτῷ. |
15 Τότε πορευθέντες οἱ Φαρισαῖοι συμβούλιον ἔλαβον ὅπως αὐτὸν παγιδεύσωσιν ἐν λόγῳ. 16 καὶ ἀποστέλλουσιν αὐτῷ τοὺς μαθητὰς αὐτῶν μετὰ τῶν Ἡρῳδιανῶν λέγοντες· διδάσκαλε, οἴδαμεν ὅτι ἀληθὴς εἶ καὶ τὴν ὁδὸν τοῦ θεοῦ ἐν ἀληθείᾳ διδάσκεις καὶ οὐ μέλει σοι περὶ οὐδενός· οὐ γὰρ βλέπεις εἰς πρόσωπον ἀνθρώπων, 17 εἰπὲ οὖν ἡμῖν τί σοι δοκεῖ· ἔξεστιν δοῦναι κῆνσον Καίσαρι ἢ οὔ; 18 γνοὺς δὲ ὁ Ἰησοῦς τὴν πονηρίαν αὐτῶν εἶπεν· τί με πειράζετε, ὑποκριταί; 19 ἐπιδείξατέ μοι τὸ νόμισμα τοῦ κήνσου. οἱ δὲ προσήνεγκαν αὐτῷ δηνάριον. 20 καὶ λέγει αὐτοῖς· τίνος ἡ εἰκὼν αὕτη καὶ ἡ ἐπιγραφή; 21 λέγουσιν αὐτῷ· Καίσαρος. τότε λέγει αὐτοῖς· ἀπόδοτε οὖν τὰ Καίσαρος Καίσαρι καὶ τὰ τοῦ θεοῦ τῷ θεῷ. 22 καὶ ἀκούσαντες ἐθαύμασαν, καὶ ἀφέντες αὐτὸν ἀπῆλθαν. |
19 Καὶ ἐζήτησαν οἱ γραμματεῖς καὶ οἱ ἀρχιερεῖς ἐπιβαλεῖν ἐπ᾽ αὐτὸν τὰς χεῖρας ἐν αὐτῇ τῇ ὥρᾳ, καὶ ἐφοβήθησαν τὸν λαόν, ἔγνωσαν γὰρ ὅτι πρὸς αὐτοὺς εἶπεν τὴν παραβολὴν ταύτην. 20 Καὶ παρατηρήσαντες ἀπέστειλαν ἐγκαθέτους ὑποκρινομένους ἑαυτοὺς δικαίους εἶναι, ἵνα ἐπιλάβωνται αὐτοῦ λόγου, ὥστε παραδοῦναι αὐτὸν τῇ ἀρχῇ καὶ τῇ ἐξουσίᾳ τοῦ ἡγεμόνος. 21 καὶ ἐπηρώτησαν αὐτὸν λέγοντες· διδάσκαλε, οἴδαμεν ὅτι ὀρθῶς λέγεις καὶ διδάσκεις καὶ οὐ λαμβάνεις πρόσωπον, ἀλλ᾽ ἐπ᾽ ἀληθείας τὴν ὁδὸν τοῦ θεοῦ διδάσκεις· 22 ἔξεστιν ἡμᾶς Καίσαρι φόρον δοῦναι ἢ οὔ; 23 Κατανοήσας δὲ αὐτῶν τὴν πανουργίαν εἶπεν πρὸς αὐτούς· 24 δείξατέ μοι δηνάριον· τίνος ἔχει εἰκόνα καὶ ἐπιγραφήν; οἱ δὲ εἶπαν· Καίσαρος. 25 ὁ δὲ εἶπεν πρὸς αὐτούς· τοίνυν ἀπόδοτε τὰ Καίσαρος Καίσαρι καὶ τὰ τοῦ θεοῦ τῷ θεῷ. 26 καὶ οὐκ ἴσχυσαν ἐπιλαβέσθαι αὐτοῦ ῥήματος ἐναντίον τοῦ λαοῦ καὶ θαυμάσαντες ἐπὶ τῇ ἀποκρίσει αὐτοῦ ἐσίγησαν. |
Mark is our earliest witness to the Synoptic tradition of Jesus’s famous “return to Caesar” teaching and the source of the other versions. In Mark’s account, the Herodians and Pharisees acknowledge that Jesus does not regard people with partiality before asking him, “Is it lawful to give kēnsos to Caesar, or not?” The verb exestin, meaning “it is lawful,” is often used to distinguish what is lawful for one person but not another, as in a letter in which Artemis of Karanis requests a legal representative “because it is not lawful [ouk exesti] for a woman to engage in a lawsuit without a legal representative” (P.Mich. VIII 507 [second/third cent. CE]; my trans.). As we have seen, women, minors, and certain high-status individuals were not legally required to pay poll taxes.
The Latin loanword kēnsos here must refer to taxes collected on the basis of census declarations, but specifically taxes collected in coin, since Jesus asks his interlocutors to bring him a denarius. As Zeichmann (2017b) has argued, in Mark’s post-70 CE context, this could refer not only to a poll tax but also to the Jewish tax.18 For the Markan author and earliest audiences, kēnsos may have encompassed the punitive tax levied on Jews. By validating the payment of census taxes, the Markan Jesus assented to an institution that normalized partiality. In this light, his clever response (“Return to Caesar the things that are Caesar’s, and to God the things that are God’s”) is cast as successful because Jesus does not show partiality that conflicted with census regulations. To those being taxed as Ioudaioi in Mark’s audience, this might not have been satisfying. Perhaps, however, some would have taken solace in Jesus’s implication that Caesar is not God, which undermines the oath taken in census declarations.19
Matthew’s version of this teaching changes very little, though it is more explicit that the denarius is the “coin of the kēnsos.” Earlier in his gospel, however, Matthean “special material” (i.e., material that only appears in the Gospel of Matthew among the gospels) primes readers to think of the kēnsos in terms of the Jewish tax (Matt 17:24–27):
When they reached Capernaum, the collectors of the didrachma came to Peter and said, “Does your teacher not pay the didrachma?” 25 He said, “Yes, he does.” And when he came home, Jesus spoke of it first, asking, “What do you think, Simon? From whom do kings of the earth take tolls or kēnsos? From their children or from others?” 26 When Peter said, “From others,” Jesus said to him, “Then the children are free. 27 However, so that we do not give offense to them, go to the sea and cast a hook; take the first fish that comes up; and when you open its mouth, you will find a coin; take that and give it to them for you and me.”
Ἐλθόντων δὲ αὐτῶν εἰς Καφαρναοὺμ προσῆλθον οἱ τὰ δίδραχμα λαμβάνοντες τῷ Πέτρῳ καὶ εἶπαν· ὁ διδάσκαλος ὑμῶν οὐ τελεῖ [τὰ] δίδραχμα; 25 λέγει· ναί. καὶ ἐλθόντα εἰς τὴν οἰκίαν προέφθασεν αὐτὸν ὁ Ἰησοῦς λέγων· τί σοι δοκεῖ, Σίμων; οἱ βασιλεῖς τῆς γῆς ἀπὸ τίνων λαμβάνουσιν τέλη ἢ κῆνσον; ἀπὸ τῶν υἱῶν αὐτῶν ἢ ἀπὸ τῶν ἀλλοτρίων; 26 εἰπόντος δέ· ἀπὸ τῶν ἀλλοτρίων, ἔφη αὐτῷ ὁ Ἰησοῦς· ἄρα γε ἐλεύθεροί εἰσιν οἱ υἱοί. 27 ἵνα δὲ μὴ σκανδαλίσωμεν αὐτούς, πορευθεὶς εἰς θάλασσαν βάλε ἄγκιστρον καὶ τὸν ἀναβάντα πρῶτον ἰχθὺν ἆρον, καὶ ἀνοίξας τὸ στόμα αὐτοῦ εὑρήσεις στατῆρα· ἐκεῖνον λαβὼν δὸς αὐτοῖς ἀντὶ ἐμοῦ καὶ σοῦ.
English translations routinely obscure the specific monetary and institutional terms used here, rendering, for instance, “didrachma” as “temple tax” when the collector asks Jesus, “Does your teacher not pay the didrachma?” (Matt 17:24). Even if a lost earlier form of this story involved the annual didrachma (half-shekel) temple tax transmitted by Jews throughout the diaspora to the Jerusalem Temple, Matthew seems to have had the imperial Jewish tax in mind with his reference to the didrachma here. Jesus goes on to ask Simon, “From whom do the kings of the earth take tolls [telē] or kēnsos? From their children or from others?” Peter’s response “From others” prompts Jesus to proclaim that the children are free.
Even though Jesus has Simon pay the tax for both of them with a statēr (two didrachmas) miraculously found in a fish’s mouth (Matt 17:27), this teaching accuses the Roman kings of the earth of extracting taxes from various “others” instead of their children—that is, Roman citizens and other high-status individuals (Carter, 2000). Matthew then proceeds to rewrite Mark’s pericope about greatness so as to emphasize that children are the greatest in the kingdom of heaven. After hearing the prohibitions on oaths in Matt 5:33–37 (another section of Matthean special material) and then this didrachma teaching, the subsequent “return to Caesar” teaching must have struck Matthew’s earliest audiences as a rather critical appraisal of census practices, even if Jesus ultimately endorsed paying taxes.
In Luke-Acts, I detect little of the ambivalence toward the census found in Mark and especially Matthew. The Gospel of Luke ties the birth of Jesus to the registration itself, the apographē, in 2:1–5:
In those days a decree went out from Caesar Augustus that all the world should be registered. 2 This was the first registration taken while Quirinius was ruling Syria. 3 All went to be registered, each to his own polis. 4 Joseph also went from the Galilee, from the polis of Nazareth, to Judea, to the polis of David called Bethlehem, because he was descended from the house and lineage of David. 5 He went to be registered with Mary, who was engaged to him and was pregnant.
Ἐγένετο δὲ ἐν ταῖς ἡμέραις ἐκείναις ἐξῆλθεν δόγμα παρὰ Καίσαρος Αὐγούστου ἀπογράφεσθαι πᾶσαν τὴν οἰκουμένην. 2 αὕτη ἀπογραφὴ πρώτη ἐγένετο ἡγεμονεύοντος τῆς Συρίας Κυρηνίου. 3 καὶ ἐπορεύοντο πάντες ἀπογράφεσθαι, ἕκαστος εἰς τὴν ἑαυτοῦ πόλιν. 4 Ἀνέβη δὲ καὶ Ἰωσὴφ ἀπὸ τῆς Γαλιλαίας ἐκ πόλεως Ναζαρὲθ εἰς τὴν Ἰουδαίαν εἰς πόλιν Δαυὶδ ἥτις καλεῖται Βηθλέεμ, διὰ τὸ εἶναι αὐτὸν ἐξ οἴκου καὶ πατριᾶς Δαυίδ, 5 ἀπογράψασθαι σὺν Μαριὰμ τῇ ἐμνηστευμένῃ αὐτῷ, οὔσῃ ἐγκύῳ.
Birth and census dating issues aside, Huebner (2019: 41) is right that Luke and his audience must have viewed the census process described as plausible. Here, it is noteworthy that it is Joseph who needs to register and that he is described as being of the “house [oikos] and lineage [patria] of David,” implying the patriarchal structure of registration by the pater familias on behalf of the household. As John Chrysostom later noted, Joseph would have needed to travel to register in Bethlehem because he was a citizen who owned property there (Natal. [PG 49, col. 351]; see Huebner, 2019: 42). Luke introduces Bethlehem as the polis of David, signifying that Joseph was a city-dwelling landowner, who may thus have been eligible for a tax discount.20 This fits with Luke’s project of elevating the socioeconomic status of Jesus and his family: in Luke, Jesus is the son of an urban landowner, not a village tektōn (i.e., a “carpenter” or “stone mason”) (Keith, 2011: 71–123).21
Luke rewrote Mark’s teaching about taxes accordingly. First, he replaced the term kēnsos with phoros, a word that could refer to the land or head tribute, but was more often used for the land tribute. Unlike Matthew, Luke does not connect the denarius directly to the tax, thereby allowing that phoros does refer to land tributes paid in kind. When Jesus is on trial, the chief priests and scribes falsely accuse him of “forbidding us to give tributes [phoroi] to Caesar” (kōlyonta phorous Kaisari didonai; Luke 23:2). Similarly, in Acts 5:37, Gamaliel’s speech insinuates that Jesus’s detractors thought of him as similar to Judas the Galilean, who “rose up in the days of the apographē.”22
Luke-Acts insists that Jesus was not an opponent of the census and related tributes. There are no hints that this author or audience had to deal with the Jewish tax. And, instead of interrogating the partiality of census practices, Luke-Acts uses the census as a literary tool to raise the socioeconomic status of Jesus’s family and to differentiate him from tax rebels and other low-status criminals (cf. Matthews, 2019).23 Luke-Acts fully participates in the normalization of the discriminatory institutional discourse represented by the census.
I have argued in this essay that New Testament scholars, myself included, have been too quick to equate the census with increased taxation and the loss of national sovereignty. Documentary sources show that the Romans did not necessarily increase tax burdens in the provinces and tended to adapt local census and taxation practices rather than impose a fully new system.
The census was an institution that regulated economic practices such as taxation and transmissions of inheritance, but it also regulated social distinctions, institutionalizing patriarchal household structures, the division between city and country, citizenship privileges, and in one important instance, the ethnoreligious category Ioudaios. Much as the US census has forced people of color like Bayoumi to identify with racial categories that misrepresent their experiences, one has to wonder how much of a role Roman censuses played in reifying social identities that were in reality more fluid. Would Christians have gradually differentiated themselves from Jews if Roman census practices did not construct Jews as rebellious ethnic others in the aftermath of the First Revolt?24 This is an impossible question to answer, of course, but I think it is a helpful reminder that censuses may play a much greater part in shaping social relations, subjectivities, and prejudices than we realize—both in the ancient world and today.