Deontic Binding: Imposed, Voluntary, and Autogenic

ABSTRACT On some current approaches, deontology is the foundation of social reality. But we do not yet have an account of the kinds of deontic structures in play. One way to approach a taxonomy of deontic kinds is to understand the ways in which they bind to an agent. There are three ways of binding deontic powers to any agent. The first two emerge from a distinction between those rights and duties forced upon an agent versus those that the agent voluntarily accepts. Within the category of voluntarily accepted deontics, however, there is an interesting subtype which, rather than being a composite of deontics assembled by sources external to the agent, is instead assembled and bound by the agent to herself. There are, then, three categories of deontic binding: imposed, voluntary, and self-bound (autogenic).


Introduction
Work in social ontology is achieving increasing prominence. Searle's work (1995Searle's work ( , 2010, for example, ignited a collection of discussions about the structure of social reality and, along with writings by Tuomela (1984Tuomela ( , 1995Tuomela ( , 2002Tuomela ( , 2007Tuomela ( , 2013, Bratman (1987Bratman ( , 1999Bratman ( , 2007Bratman ( , 2014, and Gilbert (1990Gilbert ( , 1992Gilbert ( , 2000, among many others, helped to reinvigorate work in the related fields of collective intentionality, collaborative agency, the philosophy of social science, and joint planning. Part of the motivation among researchers is the hope that this loose-knit collection of related social ontology subfields might finally deliver something like a common foundational framework for the entire social sciences, thereby giving sociology, politics, social psychology, economics, organizational behavior, legal studies, management science, and the rest of the social sciences what they have so far been sorely missing-a unified core framework for a coherent science of the social. One prominent stream of work within social ontology (e.g., Gilbert 1990;Searle 1995) is committed to what we can call, the Deontological Thesis -the thesis that the most important aspects of social reality rest on a foundation of deontology, (i.e., rights, duties, obligations, authorizations, permissions, etc.) A well-worn example will help illuminate Searle's motivation behind the Deontological Thesis. Society collectively recognizes that some person, X, counts as a minister, Y, in some context, C, and that Y has deontic powers, a collection of rights and duties. He has the right to marry two individuals to each other, but in most Western countries not more than two. He has the duty to direct a wedding in accordance with local laws that prescribe the requirements for a marriage license. And he has the duty to keep his status as a minister current. In the United States anyone with a small sum of money and the willingness to fill out a few forms can become an ordained minister. 1 A minister has the power to bring a new bit of social reality into existence simply by issuing the declaration: 'I now pronounce you husband and wife'. Doing so creates a new marriage, as if by magic. It's not magic. It's by means of the deontic power the minister possesses by virtue of his social position. If the minister stepped aside and a passing mailman or a nearby grocery store clerk stepped up and issue the same words, it wouldn't work. No marriage would come into existence.
When a marriage is created what does come into existence exactly? Deontic powers. The individuals entering wedlock immediately become the gravitational locus of a constellation of rights (positive deontic powers) and duties (negative deontic powers) that did not exist before the final syllable of the minister's declarative sentence. They have the right to tax benefits, discounted family rates at their health club, and inheritance priority. Each is subject to the duty of bearing the financial debt the other incurs under joint title. 2 And if they live in a densely-populated area of China they are currently under duty to birth no more than one child.
A status function of the form, X counts as Y in C, is a representation made possible by humans' advanced symbolic cognitive faculty which enables them to experience a world filled with such representations. Status functions are constitutive in the sense that they constitute, rather than regulate, some chunk of social reality and, most importantly, provide an individual with deontic powers.
What creates these deontic powers? For Searle, it's the declarative speech act (or at least the logical form of it) that, without exception, brings into existence status functions that have new deontic powers. "All institutional facts, and therefore all status functions, are created by speech acts of the type that in 1975 I baptized as 'Declarations' (2010, 11). 3 He says, in more detail: 'All of human institutional reality is created and maintained in existence by (representations that have the same logical form as) SF [status function] Declarations, including the cases that are not speech acts in the explicit form of Declarations' (13). Let us call this the 'Pan-Declarative Thesis'. This bold, exceptionless claim gets support from cases like marriage where, indeed, we do find a clearly-formed declarative. Think also of the manager declaring, 'You are now hired', or a political candidate declaring that you have been selected as her running mate, or a judge declaring that you are subject to a fine. Each brings into existence for you, all-at-once, a collection of deontic powers by means of a declaration.
We now have an interesting argument. If you believe the Deontological Thesis -that the most important aspects of social reality rest ultimately on deontic powers -and you believe the Pan-Declarative Thesis -that declarations are what create deontic powers -then it follows that social reality rests ultimately on the declarative. 4 Assessing Searle's Pan-Declarative Thesis demands a full treatment of its own which I shall not attempt here. I will restrict the focus to two problems with it, and the important result that emerges from an encounter with those problems. The first problem is the existence of a whole class of counter-examples -cases which cannot come into existence by means of a single declarative because the agent in question lacks the deontic power to issue a successful declarative. If correct, this problem renders the Pan-Declarative false, or at least qualified in important ways. The second problem is the difficulty in locating a declarative at the root of every case of newly created institutional social reality, a non-trivial problem.
The purpose here is not to perform a complete investigation of the Deontological Thesis or the Pan-Declarative thesis but rather to use these two problems as an entryway to a better understanding of deontics in general. An exploration of the problems reveals that such problems are virulent only in one particular category of cases, a category that can be seen as one part of a threepart taxonomy. So in what follows, I will suggest that the two Pan-Declarative problems point to the need to recognize the existence of this category and the taxonomy within which it sits. The trick to any taxonomy, of course, is formulating the criterion according to which the categories of the taxonomy are derived. And the specific trick deployed here is carving the taxonomical types by the ways the deontic powers bind to agents. Looking at how the deontics bind to agents gives us a three-part taxonomy: imposed, voluntary, and autogenic (self-determined).
One benefit of the taxonomy is that it helps us isolate and explain the third, unique, self-created group of emergent social institutions that evade the Pan-Declarative's claim that all social entities are created by means of a declarative. The Thesis misses this group because an emergent, autogenic social group isn't yet the object of widespread collective acceptance and so doesn't yet have the power derived from such acceptance to declare itself into existence, just as I don't have the power to declare myself the King of England. But the category is important. It's important because autogenic deontics form an entire class of institutions like entrepreneurial businesses, grass-roots social groups, and international crime rings that demand explanation. And autogenic deontics should strike us as shocking -that potentially enormous self-generated institutions, like (e.g., Apple), can emerge, de novo, out of nowhere, from no pre-existing deontic power and no discernable, singular, declarative.
I will mostly side-step much further discussion of Searle's (or Gilbert's, or anyone else's) theory. One's particular social ontology theoretical commitment shouldn't matter. 5 It's true that what follows will perhaps be more interesting to those already committed to the Deontological Thesis, but you need not be committed to it to appreciate the utility of the taxonomy and the importance of the autogenic category. This fascinating category leaves us with one question in the end: How exactly do autogenic institutions emerge if not from some pre-existing institutional power? A complete answer requires a more detailed treatment but I shall offer an answer by way of conclusion that foreshadows future work. The hypothesis I shall advance here is that anyone, no matter how deontically deficient in rights, has recourse to three things: promises that can be made to external people and entities; pre-existing social resources for the taking (e.g., almost anyone can sign an agreement for electricity with the local power company); and deontic potentials, 'proto'-duties and rights that have the potential to be socially reified into full-blown deontic powers.

Terminology and the Absence of a Taxonomy
A summary of the relevant terminology follows. Positive deontic powers (rights, authorizations, permissions, etc.) and the negative deontic powers (duties, responsibilities, obligations, etc.) together form the deontic powers. Collections of specific deontic powers, arranged appropriately, can form social entities or social 'structures' as I use the term. Social structures consist of deontics. The process of attaching deontic powers to an agent, group of agents, or social entity I refer to as 'deontic binding'. Deontic binding may result in a new institutional token (e.g., a marriage), a new institutional type (e.g., a new type of Governmental Cabinet), or merely a change in some existing social structure (e.g., a change of responsibilities in your job).
The term 'deontic power' is a bit awkward in that it's natural to think of only one side of the deontic powers, the positive deontic powers that fall broadly under the term 'rights', as giving the agent power in the sense that he or she has an additional ability that can be exercised optionally. Married individuals can exercise their right to a tax break or a fitness club discount but are under no obligation to do so. The duties, the negative deontic powers, impose obligations upon the agent. Since we naturally think of constraints upon behavioral freedom as a loss of power, it's counterintuitive to call duties a kind of 'power' at all. Nonetheless, I shall follow established usage here and continue to treat duties as a kind of power, albeit a 'negative' power. Deontic 'potentials' or 'proto-deontics' are 'proto' in the sense that they are not yet socially reified genuine deontics. Insofar as they have the potential to become genuine rights and duties, though, they afford an important path to them. They shall be discussed in detail later.
Are deontics the basis of all social entities? The importance of the question about of the veracity of the Deontological Thesis is clear, but we lack any typology of the kinds of deontic structures at play which might help adjudicate the answer. The various rights and duties seem to serve as the atomic units of the social world so it is reasonable to look forward to something like a Periodic Table of Deontology for them. But we do not have a set of criteria for how to build such a taxonomy, nor do we even 'have a taxonomy of status functions that I find satisfactory', to quote Searle (2006, 26). 6

Not All Acts of Social Creation Can Be Explained by the Declarative
You might believe in the importance of deontics because you believe in the Deontological Thesis, but still not believe that the declarative is the source for all deontic-rich social structures, (i.e., still not believe in the Pan-Declarative) Thesis. Two reasons for not believing the thesis were mentioned: the problem that there exists a broad class of institutional creation acts that seem to operate, contra-Searle, outside the domain of the birth-by-declaration process; and the problem of actually locating a declarative at the origin of every chunk of social reality.
Let's begin with the first problem. The declarative is responsible for many acts of social creation. For example, the declaration of a marriage, the declaration of a hiring, and the declaration of war. But some types of creation don't work this way. Wendy Kopp, for example, created a non-profit called, Teach For America, as an institution 'to promote equality and excellence in education'. But it did not arrive into existence fully-formed simply from a single God-like declarative speech act. It emerged, like most self-organized social entities, from nothing more than the germ of an idea into discussions as she began to enroll people into her vision, establish institutional partners, and eventually garner more resources as until it became what it is today -a large, well-functioning non-profit with over $300M in annual revenue that recruits teachers to help poorer, under-performing schools in the U.S. Strand by strand she built connections into the rest of relevant parts of the social world, motivating and eliciting work and contractual commitments from friends, strangers, and institutions. It was a long, slow, gradual, uphill slog of effort marked by persistence and multi-faceted work rather than a single momentous declarative that brought her institution, instantaneously and in toto, into existence.
Kopp details the events that led to Teach For America. Unsurprisingly, she already possessed traits that predisposed her to such activities -she, (e.g., led an organization as an undergraduate at Princeton University that brought student and business leaders together to discuss social issues). In the course of one of those discussions she, 'had a sudden idea: Why didn't this country have a national teacher corps of top recent college graduates who would commit two years to teach in urban and rural [low income] public schools?' (Kopp 2008) So the initial idea was a 'what if?' act of imagination (and not a declaration).
The 'declaration-in-toto' that brings about, (e.g., the all-at-once existence of a marriage), is not, it appears, a good model for a certain species of social entity creation, but is a good model for others. Contrast the gradual emergence of Teach For America with, for example, the formation of the U.S. Department of Homeland Security which does exemplify the Pan-Declarative's prediction of an en masse declarative-based creation event. The Department was created fully-formed, initially as the Office of Homeland Security upon a declaration by President George W. Bush after the events of September 11 th , 2001 and converted into a full Cabinet department with the declaration of the successful passage of the Homeland Security Act of 2002 by the Speaker of the House (and the signature of the President). That bill described a range of very detailed rights and duties backed by the full weight of the U.S. federal government. The bill took time to craft and achieve the needed consensus and, once ratified, the Cabinet took time, no doubt, to implement the office buildings, the staffing, supplies, and to sort out the protocols with the other Cabinets, etc., but much of the vast and detailed framework of deontic structures did come into existence in toto upon completion of the congressional vote culminating in the Speaker's declaration of the Department's existence. These cases support the existence of a single, root declarative, so, Searle's Pan-Declarative account works in this case. 7 The first problem suggests that there is a category of exceptions to the Pan-Declarative claim that all social institutions are always created in toto by means of a momentous declaration. The Department of Homeland Security is. Teach For America is not.

Legal Incorporation as a Defense for the Pan-Declarative
There are two objections one might raise against this first challenge. In defense of the Pan-Declarative, one objection to the challenge bites the bullet while a second weakens the thesis. The first objection stands firm in saying that there actually is one critical declaration, and in the case of Teach For America (or any other formally incorporated legal entity for that matter) that declaration is made by the Department of Corporations. In the U.S. it's quite easy to get your own corporation. Just walk over to the department, hand over about $100, invent a name -'Super Company!' -invent a number of shares of stock -perhaps an even 1 million -name the officers, then the Department declares your company's existence with a stamp of certification and your company comes into existence. Searle (2010), in a detailed treatment of incorporation, seems to suggest his commitment to this option.
But the error here lies in thinking that the essential deontic structures that make a social institution have just been created (i.e., the structures that are necessary and sufficient for its existence). They have not. It's important to make this perfectly clear. Assume that Wendy Kopp awoke one morning after a deep slumber where she dreamt of a better educational system, sat up, and promptly said out loud, 'I hereby declare the existence of an organization to improve the equality and excellence of education'. Did she just create an institution? No. This is like me declaring myself the King of England. It doesn't work.
Imagine the very next thing Kopp did was to walk down to the Department of Corporations and formally incorporate a U.S. 501(c)(3) entity so as to make a legally recognized non-profit company. With a stamp of approval, the Department of Corporations officially declares her non-profit into existence, a seemingly classic case of declaration-in-toto. Does she now have a social institution? Well, in the eyes of some governmental authorities she does. But otherwise not so much. She did fulfill the conditions required to formally incorporate which provide a strand or two of deontic power so it's not that incorporating provides no deontic powers. It provides a piece of paper that she can use to open a bank account in the name of the company, liability protection for her and her officers, and a tax category. But that's about it.
Why then would I say that she has not created the essential deontic structure? Because corporate status is not essential to her goal nor the ongoing existence of her organization. Deontically, she has little more than she had after her first empty declaration. The Department of Corporations will not take on the duties of paying her bills, training her teachers, locating institutional partners, writing contracts, recruiting donors, negotiating with inner-city schools, or renting office space. Nor will it provide her the rights to write a large paycheck to her trainers, to garner speaking engagements, to decide which rural schools will receive her corps members, nor the right to meet with the U.S. President. 8 She still has to build, piece by piece, all the various deontic structures that make a functioning institution what it is. Incorporation is neither necessary for achieving her goal (improving education), nor sufficient for it, nor necessary for the organization's survival, nor sufficient for it. It is, therefore, not essential. Discussions about this point have taught me that business scholars and practitioners find this point trivially obvious while those outside of business find it hard to believe that this small legal formality -creating a legal entity -isn't the most important critical feature of building a successful business.
The existence proof of this point lies with the millions of unincorporated businesses and philanthropic organizations since the dawn of civilization before incorporation became broadly available. 9 The point is also illustrated by all the entrepreneurial ventures that incorporated only long after they already had an established business. In such ventures, months and sometime years will often pass before the team bothers to formally incorporate. Further, in criminal organizations, of course, they never incorporate the actual business (which would be an admission of criminal activity). Why not? Because in neither case is a formal declaration of incorporation the critical piece in establishing the important, relevant deontic structures. Which ones are important, then?
The ones that aid in finding and implementing a successful business model that gets you a profit (for for-profit companies and criminal organizations), and those that facilitate your organization's goals (the social mission in the case of a non-profit).
Upon the declaration of incorporation, Kopp has only one committed person (herself) and one tiny piece of the social institution in place, in contrast to the vast panoply of activities and deontic structures necessary for a successful, self-sustaining social entity. She had, in other words, basically no deontic power. And since an institution is constituted by its deontic powers (per assumption of the Deontological Thesis) there is no social institution (except one very narrow, weak, formal strand). This strikes me as a reductio against the declarative-in-toto account of institutional creation for these kinds of cases.
Unlike the Department of Homeland Security, neither Teach For America, nor Google, nor any other autogenic social entity comes into existence fully-formed with a robust deontic web on the back of a single identifiable declarative speech act (nor its implicit logical form).
Taking stock: trying to save the Pan-Declarative by claiming that a formal declaration of incorporation is what's essential both fails to explain the creation of businesses that don't incorporate and worse, swims upstream against the Deontological Thesis -that the deontic powers are what's essential -since incorporation provides neither necessary nor sufficient deontic powers. The attempt to defend the Pan-Declarative by biting the bullet on legal incorporation does not work.

Defending the Pan-Declarative by Suggesting a Weaker Variant
A second way of defending the Pan-Declarative does so by providing a weaker, more plausible variant. Nowhere, after all, does a proponent of the Pan-Declarative claim that there must always be a single declaration, nor an in toto act of creation. There might be hundreds of smaller, less obvious declarations that gradually bring an institution into existence. Let's call the Pan-Declarative 'strong view' the claim that there must always be a singular act of Declaration which results in the creation of a social entity in toto. Let's call the 'weak view' the claim that this need not be the case; there may be a multitude of harder-to-discern declaratives and a gradual cominginto-existence.
One benefit of advancing the weak view is that, in addition to offering a way the Pan-Declarative might still be true, it also provides a reason why the second problem mentioned earlier -of locating a declarative at the root of each case of creation -exists. If there need not be a single, catalyzing declarative, but rather a set of many less momentous declarations, this would of course explain why it's so difficult to locate it -because there is none! Searle himself, in discussing an example of a wall that erodes into line of stones serving as a boundary, something with social import above and beyond its physical structure, says, 'there may not be any specific moment at which there is a speech act of Declaring, but there must be some speech act or set of speech acts and other sorts of representations . . . '. (2010, 96). This example differs from other examples he offers, each of which is a strong-form example with a clear, singular declarative that is easily identifiable and brings into existence, all at once, a new social entity (e.g., the declaration that begets a marriage). No declarative is identified in the boundary-of-stones example and it's not completely clear whether Searle, himself, would accept the weak form Pan-Declarative. As Hindriks (2011, 383), in an attempt to try and reconcile the strong and weak views, says, 'it is not very clear what exactly Searle's position on this is.' The weak form, in allowing for the gradual emergence of a social entity through many small declarations could be the perfect explanatory tool for emergent cases like Teach For America. Unfortunately though, we're offered no analysis of any such case.
In sum, we have a category of social entities that the strong form Pan-Declarative cannot explain. There is no single declarative that brings any of these such cases into existence in toto, so the strong form fails. A formal declaration of incorporation is not what brought Teach For America into existence. And, in the weak form Pan-Declarative, the problem of locating the numerous smaller declaratives seems intractably difficult. There are two options for one who still wants to retain the Pan-Declarative in the face of these issues: keep the strong form but limit its domain of application since it's clear that it cannot explain all cases of social entity creation. 10 (In this case, the Pan-Declarative Thesis becomes merely a Declarative Thesis without the purview of application indicated by 'Pan'.) The second option is to accept the weak form and offer help on the nature and location of the numerous weaker declarations. Given the strength of all the standard examples of creation-by-declaration, completely rejecting the role of the declarative seems implausible.
Fortunately, regardless of the chosen option -a bounded strong form, the weak form, or perhaps some combination -the challenges facing the Pan-Declarative have borne a fruitful result. They bring to light and point to the existence of a fascinating and somewhat ignored category of social entities. And the reason why these difficulties exist only reinforces the importance of this category and its position in the taxonomy.

Autogenic Institutions Lack a Pre-Existing External Source of Deontic Power
Why are certain cases of social creation successfully instantiated by a declarative when others are not? Because the in toto creation of a bit of social reality is made possible by virtue of a pre-existing social structure. The minister, and not the mailman, is able to create a marriage by virtue of his preexisting social role and its accompanying deontic powers. The President, and not a manager at some retail store, is able to create a new federal Executive Office. Acts of creation by members of preexisting social institutions are the most impressive, and the critical role of the declaration in such acts is clear. But it's important to note that declarations leverage pre-existing structures of power for their success. Without those pre-existing social structures the declarative's role in deontic creation is irrelevant or non-existent (as seen in my empty declaration that, 'I hereby declare myself the King of England'). And this, it seems, is the reason why self-created deontic social entities emerge only gradually, with fits and plateaus and much effort, while entities delivered from a powerful preexisting institution appear abruptly and instantaneously, as if by magic.
When Google publicly declares the formation of a new product team, their declarative speech act captivates media interest and instantly brings into existence deontic powers. When President Roosevelt issued a declaration of war against the Empire of Japan on December 8 th , 1941, the day after the attack on Pearl Harbor, it carried incredible social impact, changing the obligations of millions of people. That war effort, just like Google's product development effort, was made possible by the extensive power of the issuing social institution. When a manager declares a candidate hired or the head of the local chapter of the Independent Order of Odd Fellows declares a meeting adjourned, these people leverage the deontic powers of their positions inside the organizations which afford them the power to do such things. If the Mayor of Topeka, Kansas, instead, had declared a state of war to exist between the U.S. and Japan we would have said that he was confused about the purview of his authority (or more likely, simply 'crazy'). When President Roosevelt did the same (with congressional approval), war-time efforts were deployed across the country, boundaries of countries were re-written across the globe, and millions of lives were lost. One institution lacked the power to successfully issue such a declaration; the other had such power.
Institutional social creation works via declaration only when the institution builder is itself an established institution (or individual) with pre-existing deontic power of the appropriate kind. Kopp lacked the deontic power to simply declare a novel institution into existence the way that Roosevelt declared a state of war into existence, and her institution is one among many of its kind -those that emerge gradually, under their own power, rather than abruptly, from some wellspring of deontic power. Declaratives succeed only in a context of pre-existing deontic power and that is precisely what self-created organizations lack at their inception.

Imposed Deontic Binding Versus Accepted Deontic Binding
It was suggested that the unique class of self-created social entities sits as part of a three-part taxonomy. I shall introduce the first two parts of the taxonomy with a question: What makes oppressive social structures oppressive? It's enough here to acknowledge the obvious point that if those social structures were voluntary rather than imposed then members would simply opt out of them rather than protest. You don't have to march before the White House with a picket sign to get out from under your responsibilities that you have as an investment banker. You simply tell your boss you're quitting. That's the advantage of deontic powers that are voluntarily accepted rather than forced upon you. What makes oppressive social structures oppressive is that they are imposed rather than voluntary. (Of course, not all imposed deontics are oppressive.) One requirement for a sense of oppression is that the members of the society have no choice in the matter. Merely being born, or because you happen to be in some particular location, subjects you to an extensive variety of obligations and rights imposed upon you, whether you want them or not. I am under obligation to pay income tax to the state and federal governments and I am subject to the full compendium of criminal and civil law code, only a tiny portion of which I have the vaguest understanding. 11 I'm pretty sure I can't physically assault a stranger because I'm bored nor does society condone me hotwiring a pretty car because I like its color. There are extensive obligations I'm under at the universities where I hold positions, only some of which have been codified. I cannot convert the campus walkways into an elaborate system of water channels in order to ride my pet dolphin across campus to the cafeteria for lunch everyday-though this restriction isn't specified in my job description nor the university codes of conduct anywhere.
Recall that deontic powers include negative powers (duties) and positive powers (rights). There is a tendency to focus only on negative powers but of course rights are just as much imposed upon us whether we want them or not. It's just that obligations must be fulfilled and rights need not be exercised. I have the right to drive on public roads and expect them to be kept mostly free of snow and obstacles, and the right to pursue a career of my own choosing. I have the right to keep a dog in my house but not a full-grown lion. I have the right to over-pay my federal taxes and, in a democratic society, to state my opinion without imprisonment.
A simple way of collating deontic structures is to separate them into those imposed upon an agent versus those the agent voluntarily accepts. It's easy to overlook the voluntary nature of creation-bydeclaration, but in these cases one must first voluntarily accept the deontic powers prior to an authority figure bestowing them upon you with that declaration. For example, in the standard wedding ceremony the minister first asks each individual whether he or she takes the other to be their lawfully wedded husband/wife. The ceremony stops if one of them says, 'Well, come to think of it, no, I don't.' It is only upon confirmation of their voluntary acceptance that the minister then issues the declaration that bestows upon each a novel collection of deontic powers.
There are, of course, many other cases of voluntary deontic acceptance and it's a five-finger exercise to run through some. An offer of employment requires acceptance before you absorb the rights and duties associated with the position. Any (non-forced) contract is voluntary. All those pesky fine-print agreements you assent to when getting utility service, cable service, a rental car, or that you implicitly agree to when you buy and use products all involve voluntary acceptance. I don't read any of the endless agreements to which I assent in order to use software but I have nonetheless officially agreed to all of them when I pressed the 'I agree' button. Any kind of offer to give a lecture, run a meeting, coach a high school baseball team, or any offer involving any rights and duties requires, by virtue of the fact that it's a genuine offer (rather than a law or a penalty, e.g.), voluntary acceptance before the accompanying deontic powers are absorbed.
I should note that the distinction here does not provide an absolute taxonomical cut. It's relative to one's culture and specific context. In other words, a wedding is not always voluntary and, when not, as it turns out, we have a term for it-the 'shotgun wedding', a wedding entered into by threat of shotgun (typically the bride's father's shotgun and a pregnant bride). In many societies weddings are not voluntary although they are in mine. In those societies, the parents or tribe will arrange the wedding, often between children as young as six or seven years old. More generally, it's trivial to convert any social role from voluntary to imposed. I simply put a gun to your head and threaten you by force to 'voluntarily accept' the rights and duties that I want you to possess. A forced job under threat of violence is called 'slavery'. The mafia and other criminal organizations engage in the conversion of voluntary contracts into forced contracts. Under threat of a pair of concrete shoes worn at the bottom of the river, one will sign almost any 'voluntary' contract. What these cases have in common is a move away from what we refer to as, 'civilized behavior'. Even the most civilized of societies, however, rely on fines, punishments, and threats of physical imprisonment and/or violence and sometimes capital punishment to enforce the deontic structures that they impose upon their members. What distinguishes the uncivilized society is the degree and frequency of the conditions under which violence and such threats unfold.
So, social positions are not absolutely fixed to one category, but this fact doesn't negate the utility of the distinction. The distinction is not about where any given social role or entity permanently resides, but rather about how the deontic properties of a given social role attach, or bind, to an agent. There are two rather obvious ways they can bind to an agent: they are forced upon the agent (e.g., taxes), or they are voluntarily accepted by the agent (e.g., marriage). I take this distinction to be fairly straightforward and non-tendentious and, once you see it, you can practice applying it in various cases.
It's useful to describe a few seemingly borderline examples for purposes of illustration. Some individuals are unable, despite their best efforts, to garner the kind of, (e.g., employment offers, they find agreeable). They then view themselves as forced to accept a job, (e.g., at McDonalds serving French fries), despite the lack of any threat of overt physical violence by any punitive arm of the government. It's a legitimate question whether this person is genuinely forced, albeit forced not under a threat of violence but threat of poverty and greater degrees of ill-health and malnourishment (of the kind, e.g., one experiences in a country without much of a social safety net). Are the duties of the job imposed duties forced upon the agent, or voluntarily accepted duties? Technically they are voluntarily accepted but the question is whether this kind of acceptance is a genuine choice or whether it's more of a forced choice (like 'agreeing' to the mafioso's request with a gun to your head). It is admittedly unclear and jiggling the contextual details will sway one's intuition one way or the other (except perhaps in the eyes of a pre-committed libertarian or staunch socialist).
There are many voluntary roles that nonetheless include many unwanted duties known ahead of time by the agent. Imagine a case where Google offers me a job, I accept it, and they declare that I now retain the rights and duties of an official employee. I have voluntarily agreed to a wide collection of duties specified in my job description some of which I do not want but cannot decline without declining the offer as a whole. I don't see this case as different from almost any other case of a typical offer. I do not want to assent to the various terms and conditions in order to get cable t.v. or use some software, but I do. This kind of case does not seem borderline to me. One is capable of voluntarily accepting the conglomerate terms of an agreement and whether or not one likes all of those terms is irrelevant as to whether the agreement is voluntary or not.
Another purported borderline case is that of, (e.g., an accepted job offer) that contains (inevitably) both explicit, specified duties and unknown, unspecified duties, some of which the agent does not want. You knew you would have to write software code when you took the job at Google but you didn't know that you'd have to write code in that programming language (Java -blecht!) or work on the new Google cobalt bomb (you're a pacifist, after all), etc. The degree to which this case, like the case of the unwanted job at McDonald's, is a case of imposed deontic structures rather than voluntary is the degree to which other options actually are, or are not, freely available to the agent.
Neither such cases, nor the fact that social roles are not fixed permanently in some category, tell against the viability of the distinction as a reasonable means of culling apart imposed deontic bindings on the one hand from voluntarily accepted deontic bindings on the other. This bindingbased distinction is basic and, I would suggest, useful as an analytical lens.

Autogenic Deontic Binding
We have in hand now two categories of deontic binding -imposed and voluntary. Examples from the last section showed cases where deontic powers were derived from sources external to the agent. Take an imposed duty, (e.g., a fine for violating a parking code). It derives its force from an interlocking set of (typically governmental) authority systems that pressure you to pay the fine under threat of ever larger fines, and the loss of your right to drive. A voluntary deontic structure, a job for example, carries with it deontology derived from its parent institution. The rights and duties you possess in your role as a software engineer at Google derive from Google's vast deontological power. You have the right to gorge yourself on free food from their cafeteria because Google has the deontic power to acquire and give you that food. Similarly, you have the right to reconfigure code on their servers because Google gave you the authority to do so. The source of imposed and voluntary deontic power is controlled by entities external to the agent.
But remember that there exists a class of entities that are not created by declaration because they do not derive from an external pre-existing institution that has the deontic power to bring a new social entity into existence solely by means of a declaration. This is the category of emergent, self-created organizations: (e.g., Kopp's Teach For America), the bootstrapping entrepreneurial start-up, the emergent Columbian drug cartel, or your neighbor's local bird-watching club.
Into which side of our distinction do these peculiar autogenic cases fall? Kopp created the rights and duties requisite for building Teach For America. She took on the duty of converting her thesis into a proposal for prospective corporate sponsors, she (at a later date) assumed the right of deciding who the executive team should be, etc. Concomitant with creating such rights and duties it can be said that she voluntarily accepted them. So, these emergent social institutions with self-created deontics fall onto the side of the voluntarily accepted deontic powers rather than on the side of the imposed deontic powers. But these are different from the rest of the voluntary deontics in that their powers are controlled and structured by the agent, rather than externally.
There must be then, within the category of voluntarily accepted deontics, a further distinction to be made based upon who controls the composition of the rights and duties that are being accepted. There are deontic powers that either already exist in some institution, as a rich source of deontic powers, or are composed by the agent. To repeat, only social entities created out of pre-existing parental deontic powers can be created in toto by a declaration because only those parent structures have the requisite power to do so. Emergent, autogenic entities do not. They are forced to bootstrap their way into existence. Autogenic entities, as it were, rely only upon the deontics that can be selectively and carefully assembled.
We have, then, a three-part taxonomy of deontic powers: 1-imposed upon threat of punishment; 2-voluntarily accepted; and 3-self-assembled. Or, more briefly: imposed, voluntary, and autogenic. The first two rely upon pre-existing institutions with pre-existing deontic powers which are deployed (via imposition in the first, or voluntary acceptance in the second), while the third must bootstrap. The second and third involve deontic powers that are always voluntarily accepted, while the first forces them upon its members. And, finally, the third is a sub-category of the second. It is easiest, however, to think of all of them simply in terms of how they bind: one class imposes deontics upon the agent; one class binds only when the agent voluntarily accepts such deontics; and one class of deontics is controlled and assembled by the agent.
Another way to look at the categories is in terms of necessary and sufficient conditions for acceptance. For the imposed category, acceptance by the agent is neither a necessary nor sufficient condition. The agent's acceptance is irrelevant to the obligations imposed by, (e.g., the criminal code). For the voluntary category, acceptance is necessary but it is not sufficient because the deontic source must also make the offer. The employer must make an offer of employment for which your voluntary acceptance is a necessary condition. For the autogenic category, self-acceptance of what the agent assembles is sufficient onto itself, at least for the initial steps. Some, (e.g., Von Wright 1963, 1977, have advocated for a distinction between what they call, 'heteronomous norms' versus 'autonomous norms'. Heteronomous norms are norms that are 'given to somebody else ' (1963, 76), and autonomous norms are those that are, 'given by some agent to himself' (77). Von Wright is unsure if autonomous norms form a legitimate class and settles on the idea that it does, but only in a secondary sense, derivative from standard heteronomous norms. Although the class of autonomous norms bears resemblance to our autogenic category, it's important to make clear how his two-part distinction differs from our three-part taxonomy. First, on my view, the autogenic category forms a legitimate category. More importantly, within heteronomous norms there isn't any means of distinguishing between those that are imposed upon an agent and those the agent voluntarily accepts. Very different deontics are lumped into the single category of heteronomous norms which the heteronomous/autonomous distinction does not tease apart.

Applications of the Taxonomy
This completes the overview of the three categories of deontic binding. The taxonomy provides some benefits. It, of course, fills the gap of not having a basic taxonomy of deontological types by providing one.
It also allows a defense for someone who wants to retain the strong form of the Pan-Declarative Thesis. One option (as mentioned) is to, instead of rejecting the thesis, modify the extent of the domain to which the thesis applies (making it merely a 'Declarative Thesis'). It's not that social entities do not come into existence via in toto declarations. They often do. Social roles (e.g., an employment position) come into existence via declaration as well. But with the taxonomy in place, we now see that declarative acts of in toto creation occur only in the imposed and voluntary categories which have pre-existing external sources of deontic power that allow for such successful acts of declaration, not in the autogenic category where there is rarely enough pre-existing deontic power. The exceptions to the Pan-Declarative are not indicative that Searle's account is wrong, one might say, but that there exists an entire class of autogenic social entities beyond the purview of the Declarative Thesis. This is a clean solution, one that acknowledges the work done by the declarative, rather than rejecting it entirely, and still leaves open the possibility that the weak version might (or might not) apply in yet-to-be-determined circumstances.
What I'd like to do now is investigate a bit more this unique, third, category -the class of autogenic binding -which will help illustrate further applications of the taxonomy. Let's start with an example that describes an abrupt transition from autogenic deontics to voluntarily-accepted deontics. You decide to run for the political position of senator. This requires absorbing the relevant duties for a successful run -filing your candidacy, soliciting campaign contributions, building a marketing team, etc. In doing so, you also assume the rights to choose the marketing team and to decide what steps should be taken in the run-up to the election. Both the positive and negative deontics are self-created and self-absorbed. You take on the various responsibilities. You solicit external deontic powers in the form of money and commitments from others (e.g., a marketing agency), to absorb duties to you in exchange for some of that money. At this point the deontic powers associated with your role are self-determined (autogenic).
But once you win the election something interesting happens. Upon inauguration you immediately, per declaration of some governmental authority, absorb a wide collection of deontic powers that comprise the position you campaigned for, and these powers derive from the powers of broader governmental structures. So, an internally-determined collection of deontic powers transforms, upon inauguration, into a set of externally-derived deontic powers. Upon the voting results, you are either offered or not offered a new set of powers as part of a new position. I'm not aware of anyone, after winning, ever declining the responsibilities of office during inauguration but it's certainly possible to refuse to accept them just as it is possible to say 'no' to the minister when he asks if you take your partner to be your lawfully wedded husband/wife. The taxonomy rather easily explains this clean deontic transition as one from autogenic deontics to voluntary deontics.

The Autogenic 'Paradox'
The autogenic paradox arises from the fact that agents somehow bind deontic powers to themselves while lacking any deontic powers to bind. How can this happen? Put another way: In what sense are autogenic deontics self-generated and self-controlled? The basic answer is that an agent will start with whatever she has and selectively aggregate from there to finely control the composition of the deontic bundle needed to execute its intended purpose. The hypothesis I shall advance is that, no matter how deontically deficient in rights, status, or legitimacy, any agent has recourse to: (1) Standard promises to external parties (2) Standard social resources within reach of the agent (3) Deontic 'potentials' (or 'proto-deontics'), of two types: (a) Self-commitments that have the potential to be socially reified into full-blown social duties (b) Self-assumed behaviors that have the potential to be socially reified into full-blown social rights It's important to note that there is no doubt more to be said about the creation of autogenic deontics than discussed here, and future work in the field will, I hope, find additional alternatives. It's also important to note that the explanation of autogenics overlaps to some degree with the classic debate about how the social derives from the non-social -how a team of painters, (e.g., derives), from the intentions in the heads of participating individuals (e.g., Bratman 2014). I am under no illusion that deontic potentials are the complete solution to that deep debate, but they might provide some piece of the solution. There are further mechanics that will need to be uncovered, and some of the most interesting future work, no doubt, lies is the discovery and analysis of them.
Why should we care about the autogenic paradox? Lacking a theory for autogenic organizations leaves a lacuna; it leaves no way to explain the formation of this enormous category of entities. Kopp's Teach For America is a real organization that grew real deontic power starting from only an imaginative 'what if?' in her mind. And hers is one of thousands of institutions that grew from nothing. Like Kopp's, most of the biggest corporations began as little more than a germ of an idea -(e.g., Apple, the Virgin Group, and Microsoft). Social movements, charities, universities, book-reading clubs, governments, crime syndicates, and hospitals began, in many cases, with no pre-existing deontics. And the examples extend beyond organizations to self-made dictators, shamans, and individual business magnates. A deep survey of all possible techniques and methods that explain all examples in this diverse group would requires a wide collection of studies. The goal here is suggest a plausible set of techniques that underly all of these.
If the Deontological Thesis makes sense -that each organization is ultimately composed of a unique bundle of rights and duties (Gordon and McBride 2018), then we must face the glaring question about how it goes from zero to a unique and often enormous bundle of deontics. An answer should not only solve the autogenic paradox but fill this major gap in our socio-theoretical framework.
An autogenic agent's primary goal is to amass a large and interlocking web of deontics with the rest of the world so as to compose a social entity that can accomplish some purpose that the individual can not. A commune, a team, a government bureau, a firm, a gang, and a family business all have capabilities and powers that far exceed those of the lone individual. An appropriately constructed social entity offers capabilities and powers to accomplish things no individual can. The agent needs rights, recognized by other people and entities, to advance a budding venture, so the agents needs some others to accept duties to the venture. If we understand stakeholders as those that have an ongoing deontic relationship to an emerging social entity, then another way of framing the autogenic paradox is: How is a diverse, complex set of stakeholders, embedded in interlocking deontics, assembled from scratch, if the agent begins with little to no money or stakeholders, and little to no deontic power?
In what follows, we shall explore in more detail the answer described above.

Standard Promises
The first part of the answer is the simplest. An agent can make promises to external parties. An entrepreneur can sign a Small Business Administration promissory note for a startup loan. Wendy Kopp signed an agreement with Union Carbide's philanthropy department, promising to use their $26,000 for the purposes as described -the improvement of youth education. Promises of equity (ownership) in some firm are often made to prospective employees when a startup does not yet have money to offer actual wages. And a wide range of promises are made about the vision and plans for the social entity to a variety of stakeholders like investors, prospective board of directors, manufacturing partners, etc.
Of course there are limits to promises and limits to what kinds of promises external people or groups will believe. Those who build social entities, like startups, or social movements are famous for issuing promises at the limits of (or beyond) their ability to deliver. Bill Gates famously promised IBM an operating system for IBM's anticipated entry into the personal computer market when Gates had no team to build such an operating system, and one didn't exist. (In a fascinating turn of events he received a call out of the blue from a small computer firm, Seattle Computer Products, that coincidentally recently built such an operating system, not knowing what to do with it, and Gates crafted a negotiation that ultimately left that company bankrupt but made billions of dollars for Microsoft.) Steve Jobs promised to deliver fully complete computers to his local computer store as his first product, but only delivered extended circuit boards built in his parent's garage. Apple Computer began as a (partially) unfulfilled promise.
Promises have been discussed extensively in philosophy, linguistics, social psychology, and in more practical applications in works like Solomon and Flores (2003) and Sucher and Gupta (2021), so we shall move to the second tool of the autogenic agent.

Standard Social Resources within Reach of the Agent
An agent can, of course, use pre-existing social resources. The entrepreneur, like Kopp, relies upon anything that she can get her hands on to build her organization, including pre-existing social entities (e.g., the Department of Corporations), roles (e.g., the non-profit CEO role and it's typical deontics), cultural patterns (e.g., the ability to more easily persuade people to donate to a charitable cause than a non-charitable cause), and institutions (e.g., the institution of language to communicate, the institution of promises to make promises, etc.). The ontogenisis of autogenic deontics are in no sense free from the 'impurities' of the social. Quite the contrary. (This part of the answer is where the autogenic problem clearly diverges from the classic problem of how the social derives from the non-social.) Part of the process of carefully selecting and aggregating a 'deontic bundle' includes the careful selection of pre-existing resources like these. An analogy might help illustrate this. Imagine that our agent requires wood and walks through the forest collecting it -perhaps wood to make a fire or build a ladder. The composition of that wood -how many twigs versus branches versus logs -is controlled and determined by the agent. The composition of the fire she wants will determine the bundle she collects -more kindling for wet conditions, less for dry, more logs for a lengthy fire, no logs for a brief fire, perhaps only mid-size branches for a strong ladder. The agent decides which pieces she wants and makes choices to compose her particular bundle, (i.e., she determines which to 'bind' to herself) (autogenic deontics). Many of those pieces, like promises, or using the department of corporations to create a legal entity, were pieces already 'out there' ready to be collected and used. The kind of bundle she needs is determined by the purpose of that bundle. Contrast this with someone who walks up to our agent and offers a different particular pre-assembled bundle of wood. The agent can voluntarily accept it but has little choice about its composition (voluntary deontics). Imagine further that the local Sheriff came upon our agent and ordered her to carry his bundle of wood upon threat of imprisonment (imposed deontics). So the sense in which an autogenic agent 'creates' her own deontics is that she controls the composition of the deontics and is severely constrained both about what promises she can make and what pre-existing social resources she can utilize -only what she already possesses or can accumulate from within her reach. What choices are available to her will depend upon her role, her status, how legitimate and feasible her activities seem relative to her history in the eyes of others, among many other things. Elon Musk's suggestion to engineer the transportation of humancontaining pods through a vacuum tunnel (a 'hyperloop') in the hills of Los Angeles can be contrasted with the same suggestion made, (e.g., by a local cab driver). Musk's suggestion sparked research initiatives, media attention, and discussions by hundreds of potential stakeholders. The cab driver's same suggestion merely some interesting conversation with a customer.
So, though an autogenic agent often starts with no little to no pre-existing deontic power, she still has options. There is no requirement that she avoid social structures or begin with zero deontics. Avoiding social structures would be precisely counter-productive, and everyone as a member of society is part of all sorts of social structures and embedded in various deontics webs with many more readily available -just typically not ones that immediately help advance the agent's autogenic goal. The agent, though, might have personal money that she can use to hire employees for her nonprofit start up. She might have friends in non-profits that she can ask for advice or commitments. She might make promises in the course of applying for grants or a small business loans. She might utilize the Department of Corporations. This is the work that must be done if the agent is committed to collecting the precise composition of wood from the social forest that she believes is best, rather than accepting a pre-built bundle in the form of a, (e.g., a normal job). If she wants precise control of the deontic composition, she must assemble it herself, piece by piece.

Deontic Potentials
Not all the needed pieces can be created from blank-slate promises nor found as readily-available social resources like sticks on the ground. There is another interesting option, one that requires more explanation. An agent can manifest deontic potentials. A 'deontic potential' is a proto-duty or protoright. The use of the terms 'potential' and 'proto' are meant to indicate that they are not yet real duties or rights but have the potential to become real, socially-recognized, duties and rights in the future. They are, by definition, and intention of the agent, meant to be transitory, a bridge from the non-social to the social. (This is where the answer to the autogenic paradox overlaps with, and provides a partial answer to, the classic question about how to build the social from the non-social.) The transition from deontic potentials, like a self-imposed obligation or proto-duty, which the social world later converts into a full deontic power, has not been discussed, to my knowledge. Deontic potentials involve a two-stage process. In the first stage, they are created by an individual and so they are 'pre-bound', that is, they are bound to the individual with little social involvement (unlike a standard promise which requires an external promisee). In the second stage they are socially reified. There are many paths to social reification of deontic potentials, but I will mention six key points about the move from a deontic potential to a full, socially recognized deontic power.
The first point is that there are many behaviors without deontic potential and we need to remove these from consideration, noting why they lack such potential. For example, a commitment to secretly tapping my head three times with my finger whenever I enter my office alone is a behavioral commitment that has no potential for social reification. If no one else knows about it, nor can meaningfully engage with it or its effects, there is no potential for social interlocking or recognition of it. It is effectively isolated from the rest of the social world. (Of course, one can imagine pre-existing laws or mores that prohibit or facilitate honors and rights to people that engage in this behavior.) There is a large collection of behaviors like this one hat have little deontic potential because they don't make contact with the social world.
Second, it's clear that proto-duties possess more potential to become full-fledged deontics than proto-rights. Presuming that you have the right to issue random orders to people on the street is highly unlikely to gain social acceptance and so has little deontic potential (but, as an exception that proves the rule, cults grow on the self-assumed rights by cult leaders which followers respect, even to the point sometimes of assenting to suicide orders by the leader). Self-assumed duties, in general however, are much more likely to achieve social recognition in part because the bar for social participation is much lower. That is, the burden of work for a proto-duty, like an obligation, is on the agent, not the rest of the social world, so the agent can assume almost any proto-duty, but is radically more restricted in which proto-rights can be assumed.
Genuine rights require a pre-existing position of deontic authority. I have a right, in certain contexts, to issue orders to administrators, to receive monetary compensation for my job, to order my students to complete a term paper, etc. The difficulty for the founding agent of a new venture is that the agent often has no pre-existing power with which to facilitate the success of an order. The agent has no right to issue orders, even in her own start-up in the beginning because she, as yet, has no deontic power. But the agent can make self-commitments (sometimes backed by social promises to others about such self-commitments). So the agent, it seems, can't run around dispensing orders, but the agent can amass a great deal of proto-duties by means of establishing self-commitments and self-promises. These are self-regulated 'duties' of a sort. The general rule of thumb here is: proto-duties and duties typically come before proto-rights and rights.
Some full-fledged (socially recognized) duties can be mimicked solipsistically. I can assume the duty to maintain a vigilant watch for burglars in my neighborhood without anyone or any organization bestowing me that duty or even acknowledging it. Many other duties cannot be self-ascribed. I cannot, (e.g., assume the duty) to deliver the U.S. President's State of the Union address because that duty interlocks with various people, organizations, and institutions that would collectively prevent my speech that can only be delivered by someone holding the role of President. But almost all deontic potentials begin as self-assumed 'duties' of the kind that can be assumed solipsistically.
Third, this implies that established roles, (i.e., composite bundles of rights and duties), where the proto-duty is assumed first, provide a pre-established path toward the social recognition of the rights as well, since certain rights go hand-in-hand with certain duties. How does the agent of an autogenic institution get to the point of possessing rights, as well as duties? An agent can receive a right in exchange for a promise directly. But this isn't the usual means of amassing rights in a self-created institution. The agent typically assumes by fiat one half of the deontic powers -the duties -of a social role that is known to contain a collection of both rights and duties. Some individuals, groups, and powerful social institutions will, over time, then provisionally treat the agent as the bearer of that complete composite role containing both sides, both the duties and the rights -no longer just the duties.
Kopp's self-ascribed duty to advance youth education in the U.S. and taking upon herself the proto-duties of a non-profit CEO go along with the real rights of a CEO -to hire and fire employees, make executive decisions for the company, etc. Kopp did what the founder of every autogenic organization did. She assumed, by nothing more than fiat, the role of a non-profit director with (in this case) a mission to improve education. But she had none of the rights associated with that role. She wrote a proposal. She discussed it with anyone who would listen. She pitched her idea to potential corporate sponsors. She made a compendium of promises about what she would do if she had the requisite powers. Potential sponsors knew she wasn't in reality the director of a mature nonprofit but a few -not many -were willing to cautiously and provisionally see her as such a director based on the duties she already had absorbed that are part and parcel of such a role. Union Carbide gave her seed money and the right to use one of their offices in any manner that a real director of a real non-profit would. These were her first rights. Without the prior self-commitments that she made about Teach For America, she would not have received such rights. So an agent typically first accepts novel proto-duties which garner provisional, piecemeal recognition from others, that eventually become actual duties.
This should strikes us as nothing short of astonishing. It is not just a matter of making promises and utilizing social resources to assemble the deontic network needed. An agent can also build on self-commitments and self-ascribed 'duties' in a way that is beneficial to her social entity such that others then treat the agent as if occupying a role that, as of yet, consists of no rights at all. There is a basic element of associationistic psychology to this. The agent assumes certain responsibilities. Others then implicitly associate the real duties from a familiar social role that typically contains those responsibilities. They then cautiously and provisionally treat the agent as holding both sides of the deontic powers for that role -the rights as well. 12 Fourth, this coheres with institutional theorist's views of institutions (Meyer and Rowan, 1977;Scott, 1995;DiMaggio and Powell 1983) and especially Scott's (2014, 96) 'normative' pillar and 'cognitive-cultural' pillar activities which include roles, jobs, routines, habits, repertoires of collective action, predispositions, and scripts, all of which can provide a wellspring of 'templates' from which external agents and groups can recognize duties in some autogenic agent that are commonly associated with specific rights. The only addendum needed to Scott's stance is that such templates are often copied by autogenic agents to establish proto (not socially recognized) versions. Recognition of these composite bundles of rights and duties facilitates the move from stage-1 deontic potentials to stage-2 social reification of real deontic powers. 13 Fifth, the true test of whether a self-ascribed set of deontic potentials has been reified is whether the rights (and not just the duties) have been established. Kopp assumed the right to hire employees which was recognized, most importantly by the employees themselves. Wozniak accepted and executed work assignments from Steve Jobs. Anyone can adopt proto-duties, but it's the social recognition of proto-rights, converting them into actual rights, that really determines whether the composite role has been established. Anyone can assume certain proto-duties that can become fullfledged, socially recognized deontic duties, but rights require the deeper participation of external social agents and groups, and that deeper participation occurs in the form of duties to the agent.
Sixth, and finally, rights will often scale up in proportion to duties. Kopp has many more rights and duties now then at the initial founding. The U.S. President has more rights and duties than a State Representative.
To summarize, autogenic deontic powers derive from three things: standard promises to external parties; social resources within reach of the agent; and deontic potentials in the form of proto-duties and proto-rights. Proto-duties are more common. They transition through a two stage process where an agent assumes responsibilities or self-commitments. In the second stage those self-commitments are often recognized as duties that are part of a composite role with their associated rights. Gradually, legitimacy builds and increasingly powerful declarations can be issued on the strength of the complete deontic powers that contain both duties and rights. The process is complete when the proto-rights as well as the proto-duties have been socially recognized as real rights and duties.
I should note that there are a number of potential approaches to the paradox of the autogenic institution and, seen from the perspective of the paradox, much of the existing work on joint action, joint promises, and collaboration can be seen as suggesting possible answers to the problem. I will not adjudicate that assortment of efforts here. Further, this work intersects with an unusually large number of related fields -organizational behavior, stakeholder theory, social psychology, management, entrepreneurship, anthropology, linguistics, institutional theory, institutional entrepreneurship, and of course social ontology, collective intentionality, and social epistemology. Explicating how the current work relates to all the various streams of related research would leave no room to describe the basic theory so I must leave that discussion for future researchers.
My goal was to sketch one plausible answer to the paradox. A benefit of the approach revealed here is that it offers a framework for compatibility, rather than competition, between various fieldsentrepreneurship theory, institutional entrepreneurship theory (an addition to classic institutional theory), stakeholder theory, and small-scale collaboration theory -since these can be seen as attempting to understand the initial stages of autogenic creation. Searle's Declarative Thesis can be seen as applicable only to the categories of imposed and voluntary deontics but not the autogenic category, thereby localizing and avoiding the problem of where declarations fail. There is a reasonable collaboration and division of labor that is possible.

Conclusion
According to the Deontological Thesis, deontics are the building blocks of the social world, but so far we have lacked even a first-cut for carving the deontics. The primary goal here was to suggest a three-part taxonomy for how such deontics bind. We began with a pair of problems for the Pan-Declarative which pointed to the existence of a unique category of deontic binding. Within that category, deontic powers are self-controlled and self-assembled in the sense that the agent initiates and determines how they are accreted. The Pan-Declarative Thesis strong version works only for the first two categories -imposed deontics and voluntarily deontics (thereby rendering it merely a 'Declarative Thesis') -because only the institutions behind imposed and voluntary deontics have the requisite pre-existing power to bring a new social entity into existence merely by declaration. Autogenic, self-created social entities don't have that luxury. As such, this final category requires an explanation of the paradoxical emergence of autogenic social entities: how can agents bind deontic powers to themselves and their social entity while typically lacking deontic powers to bind? An agent must create them and this is done through social promises, existing social resources, and deontic potentials that involve a two-stage process which consists: first of adopting self-ascribed proto-duties, the real versions of which are often associated with rights that form a composite whole in some established role, like that of a CEO; and second, executing on the promises and deontic potentials until both proto-duties and proto-rights become real socially reified rights and duties.
There are many avenues for future applications. The most obvious application is the analysis of social entities in terms of their deontics and, insofar as they can be individuated, the categorization of them within the taxonomy. Future work would benefit from novel modeling frameworks that describe not just agents (as is done in, e.g., agentive modeling) but the specific rights and duties that constitute the elements of the agents and social entities. This won't do justice to some of the other strands of the social world as described by, (e.g., Berger andLuckmann 1967 or Scott 2014), but if you believe in the importance of the Deontological Thesis then you can feel assured that the most important level of analysis has being targeted for analysis.
Another application is in stakeholder theory which, to date, has suggested a panoply of ways of understanding the stakeholder. From the framework of deontic binding, a stakeholder is simply an agent or entity that has an ongoing deontic relationship with another agent or entity. An investor in a firm assumes the duty of providing funds to the firm and has rights to a financial return subject to various conditions. His role as a stakeholder ends when his investment ends. An employee is a stakeholder that assumes work duties in exchange for compensation. These are both transactive deontic relationships. Deontic relationships can also be structured under a common goal (McBride and Wuebker In press). Steve Jobs, Steve Wozniak, and Ronald Wayne established Apple under the common goal of building a successful firm. Wayne's ongoing deontic relationship ended when he cashed out early for only $800.
Perhaps the greatest immediate impact could be on entrepreneurship research which has encountered difficulty both defining its topic of study and finding the appropriate level of analysis. From the view of deontic binding, both are relatively straightforward -entrepreneurship is essentially the construction and control of the deontics that compose a social entity, purpose-built to generate economic value (or social good in the case of a non-profit). While analyzing, (e.g., the business model, product-market fit, customer validation, or design of the product are all important), analysis from the perspective of deontic binding asks a fundamental question: Does each specific right and duty and the unique, specific aggregate of them achieve the intended purpose of the constructed venture? Do we have the appropriate bundle of firewood? The Autogenic Paradox is an exploration into, among other de novo social entities, the entrepreneurial venture.
Much work remains, especially in the suggested solution to the paradox. The use of promises, existing social resources, and deontic potentials suggest one answer to it. But it is enough, for present purposes, to have made explicit the basic taxonomy and the potential theoretical utility of the taxonomy of deontic binding that includes imposed, voluntary, and autogenic deontic powers.

Notes
1. The Universal Life Church, www.themonastery.org, sells a minister's license for $20-$50 and claims to have ordained celebrities like Johnny Carson, Sharon Stone, and Conan O'Brien. 2. In the U.S., there are now couples that remain together but are dissolving their marriages to prevent medical debt obligations from spreading between them. 3. Searle capitalizes 'Declaration' as a technical term, a category of speech acts. I shall not because I find it distracting. 4. On Searle's account, deontology grounds social institutional reality, specifically. 5. . . . although Bratman (1987Bratman ( , 2014 rejects the Deontological Thesis so he, and those in his camp, will admittedly find a story of deontological types less interesting. 6. The best attempt I've seen is von Wright (1963), which will be discussed below. 7. There are a number of criticisms one might advance against this example almost all of which leverage off of the fact that the process, as described, has been over-simplified. The declarative works, one wants to say, only because of a large and complex set of structures and events that make such a declarative possible. This is true, and there's a lot that could be unpacked, but for purposes here I shall grant that the Searlean analysis of this example is basically on the mark. 8. She did send President George W. Bush a letter suggesting that he use an Executive Order to form a teaching corps the way President Kennedy formed the Peace Corps, but Bush's administration misunderstood, thinking she was applying for a job, and sent her back an employment rejection letter. 9. Variants of what we now call the 'corporation' began in ancient Rome but the current form started with the East India Company of London, followed by the Dutch East India Company and the Hudson's Bay Company, all of which were created to provide liability protection for colonial exploration and exploitation. 10. Again, for Searle, the claim is quite strong in that the declarative is intended to explain all cases of social creation (except the institution of language). 11. In a break from writing this paper I was pulled over for talking on my mobile phone while driving. My protest that I assumed that it was legal to talk using the speakerphone did not help. It is not, apparently (unless one is not holding one's phone), and my ignorance of the California vehicle code did not free me from the burden of the duty to abide by it. The officer overlooked a further violation (of which I was also unaware), that you cannot have out-of-state license plates on a car that has been in California more than 10 days -another obligation to which I was unaware I was bound. 12. This provides a new justification for those who advocate a 'fake it 'till you make it approach'. 13. It should be noted that the institutional theorists are generally looking to explain externally-derived deontics while we are looking to explain the de novo emergence of them. Classic institutional theory typically assumes one-way causation from the institution to the individual. Causation in the other direction has been a sore spot, at one point called, the 'paradox of embedded agency' (Seo and Creed 2002). Attempts to explain independent action orthogonal to the institution have been made by Selznick (1949), DiMaggio and Powell (1991), Fligstein (1997), and Battilana, Leca, and Boxenbaum (2009) but there remains a structure-action 'tension' in the field -see, (e.g., Hwang andColyvas 2020 andPackard 2021). Classic institutional theory is still mostly about how institutions shape individual behavior, not how agents can effect (much less create) institutions.