Toward a Theory of Office: Authority, Separability, Ministry, Accountability

Abstract

The ubiquity of office is rivaled only by its scholarly neglect. The stable realities and the debates and ethics attached to institutions of office are poorly reflected in political science and public administration. Offices serve as ministerial trusts (directed toward service, not to be owned, inherited or seized), they are structured by accountability institutions and ethics, and they are ineluctably relational – they exist in correspondence to other offices, those governed (who make claims upon offices), and notions of just and right. Examining public administrative offices from republican Rome through the medieval Catholic episcopacy to early modern England, I argue that institutions and ethics of office took shape that indelibly shaped American and Western public administration as we know it today. Fertile research agendas include the existence and evolution of public offices, the mechanics of their constraints upon behavior, oaths and commitment, their simultaneous embedment of obligation and authority, and rewards (fee, emolument, rent, benefice, salary).

Office appears everywhere in our world but little in our scholarship. As the locus of authority and a focal venue of obligation and responsibility defined under law and custom, office stands at the center of constitutional government in the United States and other democratic republics. Constitutions create offices and empower those holding office—whether collective (assemblies) or individual (executive)—to create, fund and constrain other offices. Constitutions and laws delegate powers not to persons but to offices. The responsibilities, rights and liabilities of those who hold office apply to persons depending in part on whether they hold office and when and how they do so. ¹ Myriad offices ranging from the U.S. President to the deputy sheriff require an oath before its occupation. And in patterns dating from the republican Roman praetor to the present, offices publicly locate (even announce) powers and responsibilities. Offices make public the venue of particular authorities and focus citizen complaints, legal obligations, and political and ethical expectations upon a position or bureau.

As ubiquitous as office remains over governmental millennia, its absence is notable in political science and public administration. In these disciplines, in public policy and in political economy, to say nothing of the sociology of organizations, office has been either ignored or stripped of its history and reduced to “position” or “job title.” Position captures something of the notion of office but surrenders its relational essence. ² Law and related fields remain the primary site of theorization about office, ³ yet even here the concept is often underdeveloped and tethered to a particular statutory or constitutional framework. The neglect of office is true almost across the social sciences and humanities. And it’s plausible to argue that in this moment, and across millennia of history, there are few concepts and institutions more important.

A career award lecture such as this obligates its recipient to a moment of gratitude for past deeds as well as a moment of circumspection for past shortcomings, and this moment is no exception. ⁴ I’m inclined to reflect on ideas I’ve considered in the past few decades, not least ideas about autonomy or organizational image and reputation. But this is a lecture about an emergent agenda of research and writing, not about what I’ve done in the past. I say this because I have neglected, as much or more than anyone in this room (or among the eventual readership of this lecture’s textual apparition) the idea and institution of office in my writings.

My review of office and offices will be historical and largely focused on the West, but key theoretical points lay at the center of my thinking.

• • Why Do (and Should) Offices Exist? Scholars of government have failed to convincingly explain why offices exist and why they, and not persons, are the principal recipients of delegated authority and responsibility. Government by office generally places power in common nouns as opposed to proper nouns. The normative angle here is why we might wish, as in the constitution of republican Rome, for a world of powers situated in and (largely) confined to offices. The positive question is why offices endure in a stable manner or why they “exist” in some equilibrium sense. This form of the question gestures to what I might call an Aldrichian sensibility (if we ask Why Parties? in some analytic sense, then we might also ask Why Offices?) ⁵

• • Relational Office. Offices amount to far more than position. From their Latinate root (officium, plural officia), they also imply obligations and expectations. They are ineluctably relational in the sense that they exist and function only in juxtaposition and correspondence to other offices, to notions of just and right, and to those whom they serve or govern.

• • Office as Ministerial Trust. From ancient to modern times, there were enduring and essential debates over office in which a millennium of writers developed something of a doctrine of office as ministry. Some but far from all of this doctrine is expressed in Weber’s notion of Herrschaft by bureaucracy. The coherence of this doctrine is questionable, but it included the following tenets:

  • ○ Offices cannot properly be owned; rather, they are “borrowed” and only for a time (Jean Bodin).

  • ○ Offices cannot properly be “seized” or “invaded”; they must be bestowed upon the holder by another, via a combination of election, appointment, or anointing (Ivo of Chartres and later canonists).

  • ○ Offices should not be driven or dominated by faction (Cicero, George Washington).

  • ○ Offices require a ministerial, not a mercenary approach. Offices cannot be defined without reference to those who rely upon them and look to them for service (Cicero, Isidore of Seville, Pope Gregory I, Ivo of Chartres, George Washington).

• • Accountable Office. Offices presuppose a notion of accountability and responsivity to other audiences. Offices are to be audited, called to provide information and reasons. Offices are to be petitioned, to receive complaints and requests for favor, excuse, redress.

• • Ejectment from Office. Holders of office can be ejected from or relieved of their occupancy of the position by virtue of the mismatch between their performance and the duties and expectations of the office. Those duties and expectations map a set of responsibilities onto the office, which bind its occupant more or less. A holder of office is the person who “takes the fall” for adverse events. An office is the place for which “the buck stops here.” Petitions and complaints can trigger the ejectment process (Carpenter and Brossard 2019).

• • Irreducibility to Role, Position, Ceremony. As idea and institution, office is reducible neither to role (James Q. Wilson) ⁶ nor to ceremonial position (Meyer and Rowan 1977). Offices are far more institutionalized and publicly legible than roles, yet their formalism is far from reducible to ceremonial conformity. As a result, regnant theories of organization in political science, economics and sociology fail to explain the phenomenon.

I will touch on these different points in this lecture without fully theorizing them. My hope is that even if none of these particulars proves compelling as argument, the topic—the institution and practice of office over centuries and, indeed, millennia—will prove of enduring value to scholars of government.

Notions of Office

Before discussing classical and medieval notions, let me acknowledge that the term “office” has a range of connotations, and that it travels with but is not reducible to “officer” and “official.” In his landmark study of the concepts and its coevolution with notions of corruption, the British historian Mark Knights (2021, 48) notes that

Office in the early modern period had a broad set of meanings.

1. (a) divine worship

2. (b) a position or post

3. (c) an abstraction of the latter

4. (d) the duty or service attached to a person’s position or employment

5. (e) a mental or bodily function—and hence also the function of excreting

6. (f) a kindness or service withheld

7. (g) a set of rooms for business, clerical or administrative work and the staff in it

Permit me to ignore the fifth of these connotations—office and orifice—and focus on the fabric linking three conceptual strands: (1) office as relational obligations (to God, to rites of worship or procedure, to the res publica), (2) office as position, then (3) office as organization.

This lecture is not primarily about the office of the presidency nor about legislative office (for thoughtful treatments, see Nelson 2014; Renan 2020; Skowronek 1992). These are indeed offices and worthy of our attention. Yet ministerial office merits attention and scholarship in its own right. This is not least because certain “administrative” positions (understood as magistrates) melded functions that were later considered “executive” or “judicial.”

Office Everywhere and Nowhere

In an address to the American Political Science Association, I begin with American constitutions, not at first the national Constitution of 1787 but the Massachusetts Constitution of 1780. The word “office” or “officer” appears 79 times in that document, where judges (chapter IV), legislators (Chapter I), and governors (Chapter II) each hold an “office” and so do subordinate executive officials such as the Secretary of the Commonwealth (Chapter II, Article III) as well as officers of the University at Cambridge, or Harvard (Chapter V). Your lecturer today is an officer of Harvard University. In every place where authority resides in the Massachusetts Constitution there is an office and usually an officer. ⁷ Each one of those offices required an oath of fealty to the Christian religion and proof that the minimum property requirement has been met by its occupant (Chapter VI, Article I).

Like other constitutions of the period, the Massachusetts Constitution of 1780 was preceded by a bill of rights. ⁸ That bill of rights had two articles that outlined the idea of accountable and temporary office.

Article V. All power residing originally in the people, and being derived from them, the several magistrates and officers of government, vested with authority, whether legislative, executive, or judicial, are their substitutes and agents, and are at all times accountable to them.

…

Article VIII. In order to prevent those, who are vested with authority, from becoming oppressors, the people have a right, at such periods and in such manner as they shall establish by their frame of government, to cause their public officers to return to private life; and to fill up vacant places by certain and regular elections and appointments.

The Massachusetts Constitution was antidemocratic in powerful ways, conceived in part as a reaction to what were considered the excesses of the revolutionary moment of 1776 and 1777, when state constitutions emasculated the executive and bestowed vast powers upon legislatures (Wood 1969). The federal Constitution of 1787 differed in many ways, not least in the idea that oaths for federal office could not demand adherence to any particular religion. Yet the U.S. Constitution, too, conceives of offices as essential vessels of power. The President “shall hold his Office during the Term of four Years” and no “Person holding an Office of Trust or Profit under the United States, shall be appointed an Elector” in the Electoral College.

The U.S. Constitution, like others, presumes that there will exist a set of administrative offices (and organizations containing them, like departments) that are not defined therein. The best example comes in Article II, Section 2, when the president’s appointment power is defined.

[H]he [the President] may require the Opinion in writing, of the principal Officer in each of the executive Departments, upon any subject relating to the Duties of their respective Offices.

The chicken-and-egg problem here is that no executive departments empowered under this constitution had yet been created when this document was drafted and debated, much less ratified into existence as the Constitution of the United States. The other key citation to administrative offices occurs later in the same section (Article II, Section 2):

[T]he Congress may be Law vest the Appointment of such inferior Officers, as they think proper, in the President alone, in the Courts of Law, or in the Heads of Departments.

Thorny questions abide here. Who qualifies as “an officer of the United States? Who qualifies as an “inferior officer”? Who qualifies as not an officer but an employed occupant of an office? ⁹

Before moving to statutes, let me note that contemporary democratic republics have constitutions that also employ the term office. The California state constitution uses office to describe its governor, its legislators, and its judges and devotes an entire Article VII to “Public Officers and Employees.” The government of France translates its Constitution for the Fifth Republic into English and therein employs the term “office” 23 times, for legislative, judicial, and executive functions. The idea of accountable office is provided for in later articles, not least Article 68-2:

Any person claiming to be a victim of a serious crime or other major offence committed by a member of the Government in the holding of his office may lodge a complaint with a petitions committee. ¹⁰

Statutes, too, center offices and in more than the traditional way. The literature on delegation from legislatures to agencies or executives is massive, beginning with Epstein and O’Halloran’s classic Delegating Powers (1999), continuing through Huber and Shipan (2002) and Gailmard and Patty (2021). The primary dynamic modeled is that of delegation to “the executive” or “the executive branch” or to “executive agencies,” and in a conceptual and positivist sense this is exactly right. Yet when delegation occurs in congressional statutes, it is not to agencies per se but to “officers” and “offices.”

Let us take the first such case in American national history, the renaming of the first executive department in the first session of the First Congress. In August 1789 Congress passed “An Act to Establish an Executive Department, to be denominated the Department of War,” and it begins as follows:

SEC 1. That there shall be an executive department to be denominated the Department of War, (a) and that there shall be a principal officer therein, to be called the Secretary for the Department of War, who shall perform and execute such duties as shall from time to time be enjoined on, or entrusted to him by the President of the United States, agreeably to the Constitution. ¹¹

A month later, President Washington signed “An Act to Establish the Treasury Department.” The stark differences between these enabling acts tell us something about the nature of delegation under the constitution as well as the notions of office and officer embedded within the departments. Treasury’s enabling act, too, began with the creation of officers.

SEC. 1. … That there shall be a Department of Treasury, in which shall be the following officers, namely: a Secretary of the Treasury, to be deemed head of the department; a Comptroller, an Auditor, a Treasurer, a Register, and an Assistant to the Secretary of the Treasury, which assistant shall be appointed by the said Secretary.

I return to these two statutes below. Their commonalities and differences are telling. Yet for the moment note that the first things that happen in the first legislative delegations in American national history are the naming of officers, to whom powers are given in the ensuing sections of the acts. ¹²

Now jump to the near present and consider a recent example—namely, the emergency authorization of the messenger RNA vaccine for COVID-19. The politics and procedure surrounding the Food and Drug Administration’s (FDA) authorization of the first such vaccine have been heavily debated, but the agent who signed off on the authorization, and who had legal authority to do so, was not the FDA Commissioner but the head of the Center for Biologics Evaluation and Research, Dr. Marion Gruber. ¹³ As Feinstein and Nou (2023) detail in their comprehensive study, this is an example of subdelegation, a widespread pattern in pharmaceutical regulation.

Consider, for instance, that Food & Drug Administration (FDA) civil servants approve new drug applications—including for COVID-19 vaccines—based on powers originally subdelegated from the Secretary of Health & Human Services to the FDA Commissioner (21 C.F.R. § 5.10). The Commissioner then subsequently subdelegated these authorities even further to career officials holding 33 distinct positions within the FDA (21 C.F.R. § 5.103).

The original authority upon which this authorization is based goes back to the Federal Food, Drug and Cosmetic Act of 1938 (see Carpenter 2010), which vests review and approval authority for new drugs in a Cabinet-level official, “the Secretary.” ¹⁴

§355. New drugs

(a) Necessity of effective approval of application

No person shall introduce or deliver for introduction into interstate commerce any new drug, unless an approval of an application filed pursuant to subsection (b) or (j) is effective with respect to such drug.

(b) Filing application; contents

(1)(A) Any person may file with the Secretary an application with respect to any drug subject to the provisions of subsection (a). Such persons shall submit to the Secretary as part of the application. ¹⁵

Yet as the statute has evolved, many of the decisions made upon new drugs and biologics are taken below the Secretary and indeed below the agency’s head, the Commissioner. In some cases “the director of the reviewing division” is the relevant office. Such powers are, in modern parlance, “subdelegated” to lower agency officials, who like the CBER director have legal authorization and responsibility for their performance.

Some basic facts emerge from these examples. Congressional statutes delegate not to agencies per se but more to offices. So too, the president does not, in the civilian realm, take executive or administrative decisions. To state an obvious point, the president does not approve drugs. He is rather to “take care” that the laws are executed. In the military realm, the President’s power as commander in chief invokes a different, more direct logic.

Finally, patterns of subdelegation again bestow authority and responsibility upon subordinate offices and officials. As Feinstein and Nou (2023) show, these patterns are remarkably widespread in the federal government.

With office everywhere, it is time to turn to its roots in European thought and practice.

Office Ancient: Plato, Cicero and the Praetor

Our modern notions of office descend from the Latin officium and magistratus and the way these were constructed by the Romans and their medieval successors. Yet the Romans were keenly aware of Greek precedents. The political theorist Melissa Lane (Lane 2023) has written a searching exploration of Greek notions of office, centering upon the idea of accountable office in Plato. Greeks of Plato’s time and for centuries after juxtaposed unaccountable kingship with “accountable office” (archē hupeuthunos) (2023, 3). Lane demonstrates accountability in the sense of constraint by other offices (2023, 4) and also by virtue of “a widespread family of procedures (generally termed euthunai) by which those subject to an officeholder’s powers were able to hold that officeholder to account.” These included audits and hearings, both of which were public (Lane 2023, 59, chap. 3).

Lane then turns to Plato, who offers a broad theorization of office and rule in his dialogues, especially the Laws. Among the many lessons that Socrates leaves his readers in the Laws is that government by office breaks down under certain kinds of polarization or partisanship (Plato, Laws (Leg. 4.715a8–b2)). When pushed to its extremes, factional government renders accountable office impossible.

When offices have become [sc. things to be] fought over (archōn perimachētōn genomenōn), the winners take over civic affairs so completely that they totally deny the losers and the losers’ descendants any share in office. Instead they live in close watch on each other to prevent anyone from coming into office who would, remembering the past wrongs, overturn the present arrangements. ¹⁶

The European and modern evolution of office owes more to Roman precedent and Roman law than to Greek. There was, by one esteemed writer’s judgment, no more important part of the Roman regime than that of office. As Han Julius Wolff remarks in his canonical Roman Law: A Historical Introduction (Wolff 1956, 27),

[t]he magistracy was the most important and most characteristic element in the Roman constitution. The word denotes the governing office; it comes from the Latin magistratus, meaning both the official and his office. In the Roman terminology the word “magistrate” includes all political officials, from the consuls down.

Roman offices included the consuls, the praetors, the censors, the aediles, and a range of other governing positions. As Machiavelli famously argued in his Discourses on the First Ten Books of Livy, the Roman Republic even created an office for exceptional moments: the dictator.

The nature of office and office-holding in Rome has been the subject of vast amounts of research (for helpful summaries consult Lintott 1997 and Lintott 2013). Yet several features are important for this writing. The first is that Roman officials governed in a world that by law and expectation made their decisions and ideas public. The head magistrates in the provinces and in the city, the praetors, took decisions that were in many ways not anticipable, in the sense that they would respond to cases brought before them. Yet on an annual basis they also announced (by edict, edico) plans and “policies” for their time in office. As Eberle (2024: 239) remarks of praetors, especially the praetor urbanus of Rome itself,

judicial edicts repeatedly used verbs in the first person singular, mostly in the future tense…Examples include iubelo (“I will order”), dabo (“I will give”), cogam (“I will force”). These personal pronouncements by magistrates were read out loud, and they were also written up and displayed publicly. ¹⁷

Recent studies and classic writings demonstrate a remarkable system of expectations and activity both strategic and normative structured around these edicts and upon praetorian behavior. When Cicero composed his famous oration Against Verres, he criticized Verres’ edict and the behavior around it as having created a bazaar where the performances of office were sold as if in a marketplace (In Verrem 2.1.104–19). “Pray recall to your memories, gentlemen,” Cicero proclaimed, “the wanton character of Verres’ administration of the law, the lack of uniformity in his decisions, the trafficking that went on” (2.1.120). ¹⁸

Cicero and his audience were quite familiar with the lobbying of office. As Eberle demonstrates (2024, 251), all sorts of Roman actors were regularly involved in trying to influence the composition of edicts. Not only were their actions and plans publicly stated and posted ex ante, praetors were in some sense also accountable and liable ex post, as Cicero’s arguments at the trial of Verres attest. While some praetors were shamed in public trials, those who were considered a model (exemplum) of good administration were declared as such by the Senate, as with Mucius Scaevola on his return from Asia in the 90s BCE (Eberle 2024, 254), roughly two decades before Cicero’s Verrine orations.

Cicero is better known to scholars of government for his dialogic reflections on the commonwealth (De Re Publica) and duties (De Officis). Cicero’s De Officis puts forth a situational ethics of sorts, discussing duties in relation to a citizen’s station and position. When Cicero turns to public office, he uses a range of terms, including the notion of officium as sphere of authority. It is to Cicero, moreover, that the idea of office as trust comes into view, especially when he inveighed against factionalism and partisanship in government (note the comparison with Plato’s Laws cited above, which Cicero had read carefully).

For the administration of the government, like the office of a trustee (sic procuratio), must be conducted for the benefit of those entrusted to one’s care, not of those to whom it is entrusted. Now, those who care for the interests of a part of the citizens and neglect another part, introduce into the civil service a dangerous element — dissension and party strife. (De Officis, I.XXV, 85)

In the republican and imperial periods of Rome, officium had not yet etymologically become “position.” It was magistratus and a range of particular titles that largely performed that function (Wolff 1956, 27). So too, the set of things entrusted to a magistrate was not defined entirely by the idea of officium. It was rather provincia, as Andrew Lintott (2013, 22–23) has detailed in his study of imperial administration, that defined “appointment” or “task” in territorial (provincial) affairs. And since the career of many a Roman politician and office holder was defined by service in the territories, these external experiences shaped Romans’ notion of government.

The republican period of Rome bequeathed one other legacy to subsequent generations of government” the idea of elected lower officials. Romans elected not only their consul but also a range of magisterial (judicial and administrative) officials “below” the consuls and external to the assemblies. In the long history of American republicanism, this gestures to an idea that has taken hold more at the state level than at the national level: the idea of the “unbundled executive” (Berry and Gersen 2008).

Office Medieval: The Bishop

The transformation of office from obligation to position occurs slowly and most of it after the Roman Empire. It is in part a legacy of the Roman Catholic Church and, in particular, of the medieval cotransformation of ecclesiastical office and secular office. Recent scholarship led by Michael Edward Moore (Moore 2011)—in A Sacred Kingdom: Bishops and the Rise of Frankish Kingship, 300-850—shows that European kings were slow to develop office-like administration after the fall of the Roman Empire. Warring kings, counts, dukes, and other rulers in postimperial Europe relied heavily upon bishops to codify Roman law and to administer much of the realm. ¹⁹ As Moore and others show, Merovingian and then Carolingian kings took up the codification of statutory law from these canons and also embedded episcopal organization deep within their monarchical states.

The earliest heavily institutionalized offices in western history are not—at least before the Ordonnance de 1445 when Valois France creates the first standing army since imperial Rome in the compagnies d’ordonnance—military positions. And they are not really those of the papacy either. They are bishops. ²⁰

The idea of the bishop (also from the Greek episcopos) stemmed from Jewish notions of priesthood and rulership, apostolic succession and the Roman provincial system of government. Christian bishops relied heavily upon notions of office and election from Judaism. In Exodus 18:21, Moses’ father-in-law counsels him to provide “rulers (שָׂרֵ֣י) (śārê) of thousands, rulers of hundreds, rulers of fifty and rulers of ten.” These rulers were not only to judge cases but receive appeals from the people. Apostolic succession consists of the Catholic doctrine that Jesus named Peter as head of His Church on earth and that popes and bishops ruled with a form of apostolic authority, viz., descended from this initial endowment. The spread of the institutional church from the fourth century onward created bishoprics in many of the places that provincial governors and praetors had been (Benson 2015, Moore 2011). ²¹ The critical difference is that, from the “classic” periods of liturgical crystallization in the fourth through sixth centuries (Baldovin 2018), the bishop is supposed to have been “demanded” by the people of his province and the approbation of the worshippers was from the fourth century onward a critical part of the bishop’s ordination (2018, 59).

This notion of popular approbation of a bishop—honored in the breach under practices of simony and regal investiture—is distinctive in Western history. Bishops were surely not elected in competitive ballots as in republican Rome, but two audiences—the people of parish/diocese and a set of comprovincial bishops—were in theory supposed to support a priest before he could become bishop. When Isidore of Seville theorized this notion of election in De officis ecclesiasticus (between 598 and 615), he gave it a meta-historical reading. The episcopate, Isidore wrote, differs from the Jewish high priesthood in that succession is based upon merit, not heredity (as in the lineage of Aaron) (De officis ecclesiasticus II, V; Knoebel 2008, 72–73). ²²

The idea that offices had to be bestowed in a certain way was a central theme of medieval debate and thinking. The bishops who ruled during the period form the Merovingian and Carolingian kings onward contended with a world in which secular rulers of monarchies and dynasties inherited their position. A key development in (republican, if not imperial) Roman office and ecclesiastical office in the medieval church is that such offices were not inherited or heritable. Of course they were sold (simony), but this very feature is continually contested (Benson 2015) and eventually itself abolished. Decades if not centuries of the evolution of canon law are taken up with this question.

A story of one of those canonists—Ivo of Chartres in what is now France—displays some of the thinking and behavior of the time. Ivo was perhaps the most influential canonist of his time and for centuries thereafter; his codifications were considered authoritative; his letters were read all over Europe (Rolker 2010), and he influenced the writer of the Norman Anonymous and the development of ordination in the English church (Kantorowicz 1957, 44). In the spring of 1092, Ivo refused to attend the marriage of Philip I—who wished to leave his wife Bertha and marry the Countess Bertrade of Anjou—until he went through a legitimate divorce. Sometime before October 1092, Philip later imprisoned Ivo at Puiset, for as much as a year. Ivo later heard that his allies back in Chartres were planning a war against the vice-count of the region and a jail break to free him. Ivo wrote his parishioners and rebuked them.

It is not appropriate that he, who came to the episcopacy without the force of arms, recover it by the force of arms; such would be the act not of a pastor [pastoris] but an invader [invasoris]. ²³

This idea that office could not be seized is a consistent one in the writings of Ivo and other canonists. It occurs in the midst of a centuries-long debate in the church over simony and investiture. Ivo followed Pope Gregory and other writings in arguing that office had to be bestowed through legitimate procedures. It could not legitimately be bought and could not even legitimately be inherited. One must come to office with some form of legitimacy, and via the choice of others, excluding the previous occupant.

For Ivo, elective appointment ruled out simony, whose defense he saw as a form of heresy. When Bishop Eudes of Normandy asked Ivo in 1094 whether a bishop or priest should be recognized by his fellow clerics after having bought the office, Ivo wrote a long letter summarizing Old and New Testament passages, patristic writings, and canon law, and he again used the language of invasion four times, borrowed partially but not fully from Pope Leo. Although “he can pass as a bishop before a number of the unknowing,” Ivo wrote, “he cannot be considered as such by those who look at him in their conscience as an invader and a heretic. Pope Leo said on this subject in a letter to Rustico, bishop of Narbonne, that: Nothing authorizes us to count among the bishops those who are neither elected by the clerics nor demanded by the people…. These words show us that episcopal honor should never be bestowed upon invaders. They are not even worthy of the clergy.” ²⁴

The elective reason for this logic, emphasizing both popular and clerical assent, might seem surprising, but it was real. There was nothing democratic or even proto-democratic about the pattern, but through much of western Europe after the fall of Rome, bishops were petitioned for by the people of a diocese and elected by an assembly of clerics and comprovincial bishops (Benson 1957, Moore 2011).

Besides Pope Leo, Ivo had at several sources for this judgment two sources for this concern about meriting office and not invading it. The first source was the episcopal councils that had been meeting throughout the early and middle medieval period (Baldovin 2018; Moore 2011). In these councils, as Moore demonstrates, the law of the medieval monarchies was created and refined. The Council of Paris had a particular admonition to this effect, warning of audacity in office. Ivo wrote to Eudes that

[t]he simple fact of believing that [whose who purchase office] are priests is already a full error. One must truly push them away [repulsione], as the canons promulgated at Paris testify: If someone has the audacity to penetrate [pervadere] this post of honor by the intervention of the king, he should never be received by the bishops of his province, who know that he has not been regularly ordained. ²⁵

Another important source was Pope Gregory the Great, author of the critical text Cura Pastoralis (590), who explicitly refers to the idea of positional office in his writings to the bishops of Gaul in 599.

Whoever tries to buy the priesthood at the price of silver seeks not the office but the title; he desires to be called a priest but is not one. ²⁶

Isidore of Seville later wrote of the bishop being ordained by the other bishops of the province as a strategy to protect against both heresy and the monopoly of rule by one man, “lest the tyrannical authority of one undertake anything against the faith of the church” (De ecclesiasticus officiis II, V, 12; Knoebel 2008, 74). ²⁷

The idea of defining office by demarcating it from what it could not be—possessed by an invader, purchased by a heretic—was a regular strategy of Ivo in his widely read and influential letters. In 1096, he upbraided the bishop Sanctio (Sanche) of Orléans, who had imprisoned a badly behaved cleric of his diocese and then, upon the cleric’s release, had the man humiliated in the town’s public square, reimprisoned, and tortured. Ivo’s rebuke to a comprovincial bishop, one he had ordained, was stinging.

Would it not be better to suffer the spoliation of your goods or even the imprisonment of your person … than to allow one of your clerics, without judgment, without condemnation on your part, to be delivered to the court where, like a thief, he is daily abused by insults and injuries? Such is not the office of a pastor but of a mercenary. ²⁸

The idea of ministerial (as contrasted with mercenary) office was a stable feature of patristic and episcopal writings in medieval Europe. The ethic had been expressed in a range of treatises on ecclesiastical office by Pope Gregory I and Isidore of Seville and made its way into canon law. (It goes without saying that ministry as ethic was violated widely, not least by Sanctio of Orléans. The ethic and practice could, and did, depart stably from one another.)

Canonists drew upon Jeremiah 23: 1–6, in which the God of Israel chastises his people’s pastors for failing to attend to the welfare of his people, vowing to raise up pastors who would replace them. Besides drawing upon the doctrine of apostolic succession, Christian bishops also drew upon the image of Jesus as shepherd, who proclaimed that he came not to be served but to serve (Matthew 20:28), who laid down his life for his sheep (John 10:11), and was filled with compassion for the lost sheep of his people (Mark 6:30–34). The customary instructions to a new bishop in Gaul recited these scriptural passages, and Ivo often used the term “pastoral office” to describe his position. Combining office with shepherd imagery, he wrote to an abbot about exercising the pastoral office (officium pastorale) in such a way as to provide a remedy for the sick and wandering sheep. ²⁹

Trust and Ministry

Medieval episcopal office-holding had thus already begun to theorize and embed two concepts and norms: the idea of office as trust and the idea of office as ministry. Political theorist Daniel Lee at Berkeley (Lee 2013, 2021) has written creatively on Jean Bodin’s notion of office, connecting it to larger discussions of sovereignty and Roman law. Bodin wrote that “office is a thing borrowed.” It is too early and too much to say that the canonists and medieval episcopate were somehow forerunners of that idea, but it is also clear that Bodin was conversing with a long tradition of writers who had reflected on the separability of office and person. (Bodin had studied in the Carmelite order early in his life; canon law was deeply familiar to him.) The idea of positional office as trust was common in the medieval Catholic Church and this fact has already been recognized in scholarship (Heintschel 1956), though I do not know whether it was theorized as commodatum.

At this point, the notion of “accountability” might seem far from our historical discussion, except that there were at least two critical audiences for the actions of medieval bishops that become clear upon a reading of their correspondence or the cartularies of their cathedral chapters. The first is that bishops regularly wrote to one another, and to popes, and popes to bishops, offering views on one another’s actions. Ivo became famous in part due to his assertive castigation of other bishops, even archbishops of France. His iconoclastic ways created as many enemies as admirers in medieval France and the Church. Yet the influence of his letters testifies to the fact that clerics and lords secular read these criticisms; they were hardly disconsonant with the times.

The second was the fact of ubiquitous petitioning in medieval Europe (Koziol 1992). As Thomas Bisson demonstrated in his magisterial work The Crisis of the Twelfth Century (Bisson 2009), anyone with lordship could expected to be petitioned in medieval Europe, and the petition-response system was such that a response was always expected. Ivo was both petitioner and petitioned throughout his episcopal career (and before that as the head of a monastery at Saint Quentin). He wrote to another bishop (Gallon of Paris) that appointment to episcopal office compelled a bishop to listen and even satisfy the demands of petitioners. ³⁰

One might wonder why in a lecture on public administration and political science, delivered to you by a scholar of American politics and North American political development, I am speaking to you of Roman law, canon law, and medieval bishops this way. I think it’s fair to say that after the fall of the Roman Empire, the first public administrators of the West were these bishops. Much of English and continental law depend upon their practices and their compilations of canon law. There were of course “bureaucrats” and “officials” in Chinese, Ottoman, and other dynasties (Barkey 1994, 2008; Wang 2022), but in terms of what shaped the modern West—and via imperial and colonial legacies, much of the rest of the planet—Roman and canonical ideas and institutions of office are really inescapable.

Office Modern: Trust, Corruption, Civil Service

The emergence of the modern state relied heavily upon the development of bureaucratic organization. One reason to note the legacy of the medieval Catholic church in this regard is that it composed a state of sorts and did so before and during the emergence of the exemplars of European nation-states (Grzymała-Busse 2023). Yet with the slow differentiation of secular and ecclesiastical powers under European monarchies and empires, concepts of office also evolved in “the state civil” as well as “the state ecclesiastical” (Condren 2006, 55).

The conceptual and theoretical development of modern office owes in part to notions of sovereignty (Lee 2021; Tuck 2016). Bodin’s idea of office as “a thing borrowed” (from commodatum in Roman law; une chose empruntée in his French versions) begs the question: borrowed from whom? If office is temporarily occupied and not owned, then to whom does it belong? One could imagine office as borrowed from the king, from the emperor, from the people, or from God. Bodin’s answer to this question is that it is the sovereign, whoever that is, that ultimately has possession of the office, and by virtue of this fact, the office holder is not the sovereign. ³¹

Bodin’s notion of borrowed office was situated in the midst of a European legal debate about the powers of lordship (Lee 2013, 2022). Bodin warned against the dangers of seigneurial rule, which functioned through the will of the lord or ruler, and contrasted it with that of public office, in which the tools and authorities of power were entrusted to the occupant of an obligated position but not permanently, and in which the office was in service to a larger entity: the sovereign state, or the modern res publica (Lee 2013, 427–28). By arguing that “offices belonged not to a person, but to the respublica,” Bodin designated a collective entity in possession and legitimation of offices. The result was “to detach office entirely from the scope of personal enrichment. It was from the state, not a person, that offices were ‘borrowed’” (2013, 427). ³²

It is essential to the development of office that there were, then, centuries-long debates about the propriety of selling and inheriting offices. (In early-modern England, one could sell the office itself or the rights to its succession, known as the “reversion.”) My impression of this rich scholarship is that we have come to know far more about intellectual debates about “office” than about the shiny objects—the particular posts—about which Bodin, Hobbes, Locke, and others argued. Aside from the abundant work of military historians and scholars of the fiscal state (Stasavage 2011), scholars will need more of the work being done by Michel Antoine (2003), Michael Braddick (2000) and, more recently, Marion Dotter (2024).

In all sorts of early modern debates about reimagining institutions, office-as-concept figures centrally in dialogue about what rule should and should not be. Offices, like republics in Benjamin Franklin’s apocryphal quip outside the Philadelphia convention of 1787, were considered nice things to have. Yet the aspirations of their occupants (actual and potential) and the demands of revenue-hungry empires meant that offices were trafficked in, systematically bartered, and bid. And those who held administrative office often stood to gain financially from its exercise by the imposition of fees in the enforcement of laws.

The possibility, nay reality, of corruption shaped our thinking of office. The term “corruption” is tossed about today in economic development studies, political economy monographs, and government accounting circles. Yet as Knights (2021) shows in a remarkable historiographical contribution that is less the study of an intellectual debate and more that of popular-legal dialogue, “corruption” surfaced as a chief concern of writers in seventeenth-century England as they began to worry that the “trust” of office was being violated not only by particular persons but by an entire system of culture and incentives. Our modern notion of office coevolved with that of corruption. In Anglo-American understanding, the two concepts are inseparable. ³³

Knights documents a world in which a language of the public good evolved alongside, and later in opposition to, a language of interest. Some politicians were, in the late sixteenth century, believed to be “public men” answerable to a larger commons. As fiduciary law developed in early-modern England, the idea of an office as trust emerged (2021, chap 5). Then “over the seventeenth century the idea of a public official accountable to the public took firmer hold” (50). Knights traces the ideas and their linkages through pamphlets and legal cases. By the early nineteenth century, it was a common idea that public officers had obligations facing toward the public that outweighed any potential responsibility to the King (who made appointments to the civil list) or even to Parliament. A 1783 case—Rex v. Bembridge (Knights 2021, 54–59)—established that even a private contractor serving as a military paymaster had primary responsibilities to the public, and was legally responsible for “the faithful discharge of his duty.” The prosecutors of Lord Melville declared in 1806 that “the duty of every public officer appointed by the King is a public duty.”

Early modern European rulers learned that two dynamics posed grave risks of systemic corruption in public office. The first was the practice of venality, which became more common, not less, in the early modern period in both England and France (Doyle 1996; Knights 2021, chap. 11). Venality exploded along with the revenue needs of the state and as a means for monarchs and lords to purchase favor from one another and from the gentry. As Whigs and American colonial revolutionaries commonly argued in the eighteenth century, its ubiquity undermined the distance of the English Commons from the royalty, rendering a mockery Montesquieu’s putative separation of legislative from executive in the English constitution he admired for that reason. Well after the American Revolution and deep into the nineteenth century, however, venality continued in England. And even after direct sale of offices ended in a series of enactments from an 1809 statute to the 1832 Reform Act, patronage persisted. In a sense, partisan trafficking in office merely replaced personal and seigneurial commerce, and it was this realization that inspired the Northcote-Trevelyan Report of 1854. In France, it was none other than Jean-Baptiste Colbert who tried—successfully, for a time—to rein in the number of offices subject to vénalité (Doyle 1996; Soll 2009) as part of state rationalization.

The second concern of English politicians and reformers was fee-based remuneration (Knights 2021, 405–8). Those who occupied English administrative offices derived a large part, often most, of their income from means other than their salary (Knights 2021, 4–11). Even so venerable an institution as the Exchequer continued allowing its officials to make money by means of fees until 1848. Meanwhile, colonial offices such as those in the East India Company were characterized by low salaries and a range of mechanisms: “fees, perquisites, favours and … permission to engage in trade.” The equilibrium of low-salary and high fees left Lord Cornwallis lamenting a false economy in His Majesty’s civil service, while Edmund Burke would describe the East India Company “a state in disguise of a merchant; it is a great public office in the disguise of a counting-house.” As Nicholas Parrillo has detailed in his careful and innovative study of fee-based governance (2013), these concerns echoed in early American government as well. The more-than-century-long replacement of fees by a salary system marked a critical feature of American state development.

As “corruption” became the concept under which these concerns were united, officials and reformers worried about the degree to which office as “trust” was being abused. The idea of trust as a concept linked to office emerges in England in the late sixteenth and early seventeenth century (Knights 2021, 111–19). The idea was especially pronounced in Leveller writings, for whom “Trustee” and “State Officer” became nearly synonymous and exchangeable—the idea of office as trust connotating a responsibility to the public, in whose stead and for whose interest. “All rule is fiduciarie,” declared the parliamentarian lawyer Henry Parker. As the duties of the office were public, critics of venality worried that partiality would be shown to fellow lords and that the possibility of ejectment for poor behavior was limited. And governance by fees induced the holder of office to abuse their powers, drawing officials’ concern away from the public and toward the next opportunity for marginal enrichment.

Parliamentarians and Levellers also conceived of offices as relational and accountable. As Knights relates his writings, the Cromwell critic and Leveller sympathizer John Streater worried in 1653 about “covetousness in a person of great trust” and considered a range of solutions to the problem. The general idea was to make public office “lesse profitable” and to consider making offices elective. Yet two other mechanisms were audits and prohibitions, particularly that (1) “it may be made Criminal for any person in publick trust to receive any Bribes, Fees, of Gifts, that shall have Causes or Petitions depending before them” and (2) that “all persons upon the resigning his or their Office, required to give a publick account; for in private accounts there may be connivance” (Knights 2021, 124). Streater’s notion of prohibitions upon fees and emoluments was premised on the relational assumption that offices of all kinds would have “Causes or Petitions” before them.

Office in Early American Government

Notions of trust and fiduciary duty are critical for understanding early American office. Before we begin to separate “presidential” from “ministerial” or “administrative” office, we might note that the first President, George Washington—upon whose personality Article II is said to be partially designed (Elkins and McKitrick 1995)—thought in terms of “office” for his entire adult life.

Before becoming president in 1789, George Washington had a long career that spanned many offices, and his later writings are shot through with the language of “office” and “trust.” ³⁴ Washington had Cicero’s De Officis in his personal library, and the language of seventeenth- and eighteenth-century English office suffused much of his thinking. Washington served as Commander of the Virginia Regiment and, later, as Commander of the Continental Army.

Washington experienced his offices as relational and ministerial. As I have written elsewhere (Carpenter 2021, chap. 2), Washington and other American statesmen lived in a world suffused by petitioning and petitions. Petitions were not merely sent to assemblies and parliaments or even kings. They were commonly sent to anyone holding an office. They were, moreover, and expected to be met with a response from that office holder. As early as 1756 Washington wrote of the “supplicating tears of the women; and moving petitions from the men, [which] melt me into such deadly sorrow.” ³⁵ Washington was petitioned at every stage of his career, not merely by appointment seekers but by widows and their representatives, by Native Americans, by current and former soldiers, and by job seekers.

Washington’s notion of office included the very ideas of ‘trust” and ministry that had been a hallmark of Anglo-American dialogue from the late sixteenth century onward. As he resigned his leadership of the Continental Army in 1783, he wrote a circular letter to the various governors of the new states, stating that the “great object, for which I had the honor to hold an Appointment in the service of my Country being accomplished, I am now preparing to resign it into the hands of Congress.” Washington’s resignation was, in this sense, a transfer of the office back to its sovereign owner, the national legislature. This notion of office as a thing borrowed informed his description of having “surrendered up my Public trust to those who committed it to me” and “a last farewell to the cares of Office and all the employments of public life.” ³⁶

Washington’s experience and thinking are reflected in early American agencies and their creation. ³⁷ Let us return to two executive departments of the Founding, War and Treasury. The first agency created by the U.S. Congress was the Department of Foreign Affairs, but it was quickly renamed, in the August 1789 “An Act to Establish an Executive Department, to be denominated the Department of War.” The War Department was known less by its organization than by its principal officer. Given the particular military powers bestowed upon the president under Article II of the Constitution, much of what the Secretary of War would do was defined by the president. ³⁸ The second section of the same act created yet another officer—“an inferior officer, to be appointed by the said principal officer, to be employed therein as he shall deem proper, and to be called the chief clerk in the department of war”—for the keeping of records. Thus, the first delegation actions of the First Congress were less to create any sort of organization for the War Department than to establish its two primary officers, one of them not to be appointed by the president.

The next clause contained oaths and introduced the term “trust.”

SEC. 3. And be it further enacted, That the said principal officer, and every other person to be appointed or employed in the said department, shall, before he enters on the execution of his office or employment, take an oath or affirmation well and faithfully to execute the trust commited [sic] to him.

Not only the Secretary and the chief clerk but “every other person to be appointed or employed in the said department” was required to take an oath to “well and faithfully to execute the trust” bestowed upon him. Oaths would be required of employees as well as appointed officials. The trust committed to the oath taker was not defined in the statute.

A final section designated the Secretary of War as the chief record keeper of the Department, “entitled to have the custody and charge of all records, books and papers in the office of Secretary for the department of war.” Here the president was not mentioned as the source of record-keeping powers. Those powers presumably derive from the statute itself.

A month later President Washington signed “An Act to Establish the Treasury Department.” The stark differences between these enabling acts tell us something about the nature of delegation under the constitution as well as the notions of office and officer embedded within the departments. Treasury’s enabling act, too, began with the creation of officers (see Section 1 quoted above).

The first difference between Treasury and War was the greater specificity in officers created directly in statute. The second was the far greater specificity of duties of the officers structured not as the president would define, suggest or communicate them, but in statute.

SEC. 2. [I]t shall be the duty of the Secretary of the Treasury to digest and prepare plans for the improvement and management of the revenue, and for the support of public credit; to prepare and report estimates of the public revenue, and the public expenditures; to superintend the collection of the revenue; to decide on the forms of keeping and stating accounts and making re turns, and to grant under the limitations herein established, or to be hereafter provided, all warrants for monies to be issued from the Treasury, in pursuance of appropriations by law; to execute such services relative to the sale of the lands belonging to the United States, as may be by law required of him;(a) to make report, and give information to either branch of the legislature, in person or in writing (as he may be required), respecting all matters referred to him by the Senate or House of Representatives, or which shall appertain to his office; and generally to perform all such services relative to the finances, as he shall be directed to perform.

Two other notes about the Treasury enabling act of 1789 merit attention. The first is the absence of oaths. Unlike the War Department where every single person working in the agency was required to take an oath, oaths are entirely absent from the initial authorizing legislation for Treasury. Second, the final sections of the 1789 Treasury enabling act turn to what anyone holding an office in the Treasury Department cannot have: competing interests in assets, income, “any business,” or “emoluments.” Section 8 held that

no person appointed to any office instituted by this act, shall directly or indirectly be concerned or interested in carrying on the business of trade or commerce, or be owner in whole or in part of any sea-vessel, or purchase by himself, or another in trust for him, any public lands or other public property, or be concerned in the purchase or disposal of any public securities of any State, or of the United States, or take or apply to his own use, any emolument or gain for negotiating or transacting any business in the said department, other than what shall be allowed by law; and if any person shall offend against any of the prohibitions of this act, he shall be deemed guilty of a high misdemeanor, and forfeit to the United States the penalty of three thousand dollars, and shall upon conviction be removed from office, and forever thereafter incapable of holding any office under the United States.

For the First Congress, the executive departments were variably structured holding tanks for “offices” and “officers.” This may well tell us something about how constitutional meaning was perceived in the Early Republic, but it also tells us about how agencies are created when there are few institutional precedents and when the flesh and sinew that completes the skeleton will come later. The second thing that leaps off the page are the various formal obligations attached to these offices—oaths (War) or restrictions on employment or income (Treasury)—as well as less formalized notions of “duty” and “trust” that are to inspire and constrain administrative behavior.

A final note about early American administration comes from President Washington’s farewell address. The Federalists’ heterogenous writings warning Americans against the dangers and mischiefs of faction have been well and duly catalogued (Federalist 9, 10). Yet Washington’s farewell address stands out in its concern for “the State” and for “public administration.”

I have already intimated to you the danger of parties in the State… .Let me now take a more comprehensive view, and warn you in the most solemn manner against the baneful effects of the spirit of party generally… the common and continual mischiefs of the spirit of party are sufficient to make it the interest and duty of a wise people to discourage and restrain it.

It serves always to distract the public councils and enfeeble the public administration. It agitates the community with ill-founded jealousies and false alarms, kindles the animosity of one part against another, foments occasionally riot and insurrection. It opens the door to foreign influence and corruption, which finds a facilitated access to the government itself through the channels of party passions. Thus the policy and the will of one country are subjected to the policy and will of another. ³⁹

Against Faction: The Long Road to Merit and Salary

Concerns about corruption in office led to what I would regard as two central developments in the emergence of the modern Weberian state: (1) the development of merit-based civil service systems (Knights 2021; Silberman 1993) and (2) the development of salary systems (Parrillo 2013). Parrillo’s scholarship is particularly important here in that it points to work at the state level (chap. 5) as well as to developments in the imperial American state (chap. 9).

The problem of patronage was not merely that it reeked of “inefficiency.” It surely did. Yet as Stephen Skowronek so elegantly detailed (1982), it also raised vast concerns about moral and ethical corruption. It was not merely that presidential parties were appointing tens of thousands of public officers, but that those same officers were highly incentivized, essentially required, to contribute part of their pay to the party purse. Office was for sale once again, but mediated through the institutions of party in which state organizations were particularly forceful. And the arguments against the patronage regime, led by reformers like Dorman Eaton, invoked not only late nineteenth-century Victorian morality but also the commonwealth tradition of republicanism, the school of thought that reacted so anxiously, and later vituperatively, to the undermining of Parliament by the patronage of the Hohenzollern kings (see also Knights 2021, chap. 13).

As with the bricolage implementation of federal civil service reform (Skowronek 1982, chap. 6), the “salarization” of American state, local, and federal governments happened piece by piece. It varied heterogeneously by government function, by locale and by position. There is direct and compelling evidence that the move from fees, bounties, and salaries changed behavior, as for instance in the drop of prosecutions by U.S. Attorneys as their remuneration was changed from prosecution-based fees to fixed annual salary (Parrillo 2013, 273–89). Parrillo also notes that, in the Anglo-American world, salarization was not a uniform development. British naval officers were permitted to pool their prize money during the First World War, whereas American naval officers, following an 1899 statute, were unable to do so (2013: chap. 9).

I conclude this essay just weeks after the second inauguration of President Donald J. Trump. It has been my wish here (and in classes I teach) not to reduce the complexity of what we study to the all-consuming and polarizing axes of his person. Yet the Trump administration seems intent on resurrecting the Jacksonian model, at least in part, and it will be able to do so in a world that presents quite different structures than it did the height of nineteenth-century American patronage. The American state is far more centralized and more easily coordinated than ever before, with powerful domestic security agencies as well as a powerful and centralized military. The mediating institutions that held back the nineteenth-century and even twentieth-century state have been weakened, in some cases fully jettisoned (Skowronek 1982). In other ways, the Trump administration’s Schedule F represents a dramatic expansion of previous attempts to “thicken” American government by creating more and more political appointees (Light 1999).

Yet along the way, two other developments have begun to problematize the notion of public office and in remarkably different ways. ⁴⁰ The first is the now ubiquitous phenomenon of public employee unions, especially at the state and local level in the United States (Anzia and Moe 2015). The degree to which they render public office less as a thing bestowed than as possessed by those who benefit from public union protections needs further investigation (for a striking argument, see Howard 2023). The second is rule through contractors (Light 1999; Potter 2022). Here the undermining of accountable office and nonpurchasable office comes through a system that modern American government has signed onto wholly.

Conclusion—An Agenda for Scholarship

As a sketch of a problem and a call for further scholarship, my lecture has alighted on many themes, moments, institutions, and concepts that have been studied in far greater depth elsewhere and/or which merit intensive study in the decades ahead. Here are the principal questions I see confronting political scientists, public administration scholars, and social scientists generally about offices and their evolution.

Durable Traits of Office

Offices have been with us for millennia. To draw any line between ancient and medieval office and modern office risks, to be sure, is a form of anachronism. Yet a first theoretical agenda might be to ask what links them, if anything.

• • The differences must surely abound, but are there meaningful commonalities between the Roman praetor, the medieval bishop, the early American Secretary of War or Comptroller, the medical reviewer for a new drug submission, and offices of the last century in American government?

• • Are there appreciable commonalities between the way that contemporary audiences conceived of these offices?

• • How have governmental systems and organizations, whether democratic-republican, ecclesiastical, patrimonial, autocratic, or other, tried to render offices accountable?

• • Why do offices seem to occupy a focal function, announcing to the world that “here is where a particular problem is to be addressed,” or “here is the agent with whom you can work to address a particular concern”? How does this relate to state legibility (Lee and Zhang 2017)? This focal nature of office seems weakly addressed by Weberian dynamics, by network theories of organization (Padgett and Powell 2012; Powell 2003) and by ceremonial theories of organization that emphasize loose coupling (Meyer and Rowan 1977).

Transformations of Office

Any realistic research agenda would also need to focus on how office has been transformed. In the area of merit-based qualification for government office, especially in terms of legislation, scholars have spilt considerable ink (Skowronek 1982, Silberman 1993). In the area of salary-based renumeration, with the shining exception of Parrillo (2013) and parts of Knights (2021), they have not.

• • How were offices created? By rule, by decree, by statute, or by other means? What logics—strategic, social, cultural—“explain” their existence and durability?

• • Outside of the American context and even within it (Parrillo 2013), how and how variably have public offices been converted to a salary basis for payment?

• • How have forms of auditing and accounting been transformed, and how have these changed altered public behavior and administrative behavior?

• • In the spirit of organizational emergence (Padgett and Powell 2012), how have offices spawned organizations and how have organizations spawned offices? The matter is far more complex than can be accounted for by organizational ecology. What aspects of formalized organization, even Weberian organization, are not reducible to office?

• • Why have some offices required oaths upon their occupation, while others have not? How have oaths evolved?

• • Which later ideas of office borrowed from earlier models, and which earlier models? The diffusion of office concepts and institutions as a general matter seems ripe for investigation.

Behavioral Office

A final set of questions might concern how offices do (or have) shaped administrative behavior. At some level, this is the question of Simon and many progenitors. Yet recent scholarship suggests that conceptions of duty are every bit as important as material incentive in civil service reform (Williams 2025).

• • Is duty reducible to public service motivation (Moynihan and Pandey 2007), and vice versa?

• • How are obligations of office conceived (Dubois 2016; Zacka 2017)?

• • How are the clients or public conceived (Fountain 2001)? This is in part the question that Washington and Cicero were asking when they worried about factional administration.

• • How do institutions of office promote or inhibit collaborative governance (Ansell and Gash 2008, 2012)? Do notions of office as duty constrain public agents to operate in silos, or might they be (have they been) more expansively conceived to promote? Or might collaborative governance need to shed offices altogether?

Government by office is a remarkably widespread phenomenon and so, perhaps, is capitalism by office. Their evolution cries out for better explanation, and at that, well beyond the Weberian, purely delegative or ceremonial models that have served us so well. As we ask whether contemporary offices merely ceremonially layer earlier institutions or embed their legacies explicitly and implicitly, we begin to address the critical interplay of ideas and institutions, in a scholarship that demands both granular description as well as robust theorization.