Effective Voice Frey Schneider
Speaker 1: So welcome. This is a last minute improvised presentation on a work in progress that Seth and I are working on. I'm Nathan Schneider. This is Seth Frey. And we are we've been this is one of the collaborations that's emerged out of the meta governance project. And we just started by reading Albert Hirschman's Exit Voice and Loyalty together a bunch of months ago and tried to see...
Top Keywords
- voice 0.021
- exit 0.021
- affective 0.013
- exit voice 0.012
- effective 0.010
- effective voice 0.009
- hirschman 0.009
- mechanisms 0.006
- internet 0.005
- loyalty 0.005
- much 0.004
- well 0.004
Transcript
So welcome. This is a last minute improvised presentation on a work in progress that Seth and I are working on. I'm Nathan Schneider. This is Seth Frey. And we are we've been this is one of the collaborations that's emerged out of the meta governance project. And we just started by reading Albert Hirschman's Exit Voice and Loyalty together a bunch of months ago and tried to see whether we could whether we could build on it. It just felt like a book that's really deeply related to the issues that that we're concerned about. I know in the last couple papers I've written, the concepts of exit and voice have felt have been relevant for me and, you know, keep coming up. And yet it is a book that was published in 1970 and and, doesn't address the Internet and doesn't address some of the dynamics that we see in in online governance. And so we wanted to, to build on that in some way and we've been wrestling with with how to do that. Seth, did you have any thoughts about how to how to organize this, conversation? You know, we could lay out some of the, you know, just map out kind of the the progress of the paper, and then then open it up.
Yeah. Yeah. Well, I I can speak to my own motivation and interest. Again, the the Internet changes a lot. It changes the physics of resources in a way that gives us a lot more freedom, to design institutions in the most abstract terms, but to define goods, divine resources, which which in turn has a cascading effect on how we govern our lives. And so when we when we looked back during our big ass lit review in the applications of exit and voice to, the Internet. You know, we found applications of exit and voice to the Internet, but a lot of them, had the appropriate humility to kinda take the framework as complete and apply it in a to, you know, usually online consumerism. We we kind of went wanted to go further, exercise our hubris, and see what what holes are poked in the framework by the affordances of the Internet or what kind of new directions can go, what assumptions of the analysis are are broken or undermined, or how we can use it as a launching pad to push our own agenda for the for for the online domain. And so that really moved us for a lot of this. Nathan, I'm I'm thinking of this this improv game where someone starts a story and then basically digs the next person clockwise into a big hole, and then that person that person digs the next person in the hole. Is that what we're ending up with, the sort of, random baton passing?
I think so. And I and and we have at least the, the the cheat sheet of the outline of of our paper. I think we could just kind of alternate and and walk through some of the the contentions we make. And and, I think it'd be useful to start out too just with a general, a few remarks about Exit Voice and Loyalty itself and and what it's about. The it's a book by a and a kind of quasi heterodox economist who, you know, really interesting figure who was a refugee from from Germany, bought in on several in several armies before and during World War two, you know, kind of polymath type figure who was trained as an economist eventually, but was had kind of a humanist humanistic orientation. And and, you know, in the book and later describes it really as an attempt to get economists to care about voice, to recognizing that economists are often very interested in these kind of transactional relationships where where the logic of exit, which is to say, you know, taking your money elsewhere when you're dissatisfied with something is the kind of preeminent form of agency that economists are trained to see and calculate. And, he's interested in this other dynamic of human activity, which is voice, which is the ability of people to, speak to and inner inner eject into, the the institutions of their lives. So he looks especially at, you know, firms, and consumer relationships. These are all very hierarchical relationships, but also things like people's relationships political parties. You know, one case he looks at is the Nigerian railways. And, he points out over and over the ways in which voice plays an instrumental role in the shaping of these organizations, and and especially its relationship to loyalty, generating loyalty as a means of, enabling kind of governance and stability and and responses to, to to tensions and problems in the organization. And it's also important to emphasize that he everywhere sees an interrelation between exit and voice. So these principles are broad principles he's seeing everywhere, but they never stand alone. So, or if they do, it often becomes problematic. So so exit can be a, you know, for instance, he he considers what at that time was just an abstract proposal by Milton Friedman now as a reality of school choice. And and explores how the introduction of certain exit dynamics, could be kind of steam valves for, dynamics that are otherwise governed by voice. You know, if if for instance, the wealthy, people in a in a public school can leave, it did it reduces the amount of effect that, the voice of parents can have collectively. And, you know, it changes the nature of the voice that remains in the public school when the, when when the wealthy parents have left. Left. So over and over looks at these kinds of relationships and how certain, you know, voice mechanisms and exit mechanisms can, play together. Is there something you want to add to that summary? Maybe in a with some Hirschman plus added.
Yeah. So, you know, the the the big epiphany ends up sounding kind of obvious. The, oh, well, people don't just treat all their interactions like market interactions and leave whenever they want. Surprise. People talk to each other and, like, and and try to work things out. And you can use this sort of human quality to explain a lot of what looks like stickiness in transactions that should be, you know, converging quickly on equilibrium. This ends up being a sort of parallel to a figure that a lot of us here are a fan of, Elnor Ostrom, who can also kind of be reduced to the same epiphany. That rather than, in the case of a tragedy of the commons, rather than, you know, dictatorship or the market, there turns out to be all this stuff in the middle. We're surprised people can talk to each other and look at the problem and come to something in common. So that's one level you could say that they are saying the same obvious thing in a way that's profound largely because of the sort of resistance of the economic community, at least in those days, to to models outside of of exit exit exit. So the reception among economists of exit and voice was exit voice loyalty wasn't great. The most friendly of them sort of said, if you're the type of economist who's not a real economist and cares about other things, you'll like this book. That was one one of the reviews we stumbled on. A more typical review said it's not that great analysis of an unimportant problem. And yeah. But the book ended up taking off in every other discipline of social science. So where Hirschman wanted it to be opening economists to the open to the idea of voice, it ended up opening political scientists to the idea of exit, and, it ended up getting a life of its own in social psychology by some people who study relationships, in labor and like urban studies of people who study why workers stay with firms, in migration studies, why people, leave or stay in countries. And obviously a ton of consumerist literature, like market research consumer lit, and there's oh, yeah. Municipalities. So this ended up tying into some arguments in the study of cities and why people move from city to city. So it really got a life of its own, and it certainly has a life of its own. I think, Nathan, I'll take this point if we're ready to transition into the Internet. Is that premature?
No. That that's Okay. Fine. Yeah. There's a kind of growing literature on on this stuff in in relation to the Internet. Much of it, we found, you know, as Seth said earlier, just kind of takes the the framework as is and applies it. And and, you know, much of this, you know, we've seen for instances about, like, how oh, you can have so much more voice on the Internet because shit goes viral and things like that. And, you know, and there are some really good examinations as as well. But, we wanted to highlight some of the ways in which the Internet changes some of these dynamics. And there, there are two well, first of all, I should just force foreshadow what we, what our major intervention I just wanna highlight this so you know what we're what we're doing with all this is to make a distinction within this concept of voice, which is a a distinction that, you know, is really central to to the kinds of things we're trying to do in the meta governance project. You know, because we're we're trying to figure out how do we how do we describe what we're talking about around these, like, governance mechanisms, in in the context of exit and voice. And what we found is the need to make a distinction between what we call you know, this is super, like, you know, gross academic stuff, but affective with an a, voice, and effective with an e. So affective voice is largely what Hirschman was concerned with, and it's largely what happens on the Internet as we understand it. Affective, you know, drawing on this tradition of affect where, you know, it's largely emotional in character. It's it can be very powerful, but it still relies on it doesn't produce direct effects. Whereas the effective voice we're talking about is something like is something like being the admin and having the power to ban somebody. Right? Whereas affective voice is a group of users in a community banding together and, you know, changing their their user profile picture altogether to the same thing. It creates a powerful emotional effect, but it doesn't actually ban somebody. Right? It doesn't actually, create a a a, you know, a kind of binary, effect on on how the system is functioning. And we functioning. And we and we we wanna recognize that there is voice that can have those kinds of effects that can be binding on the on the political structure, on the social structure in the way that, you know, voting an election or something like that is, you know, has a kind of binding way of it is a binding way of expressing voice. In the Internet context, we identify two main ways in which the the dynamics of exit and voice have are are functioning really differently than the cases that Hirschman is looking at. One is the radically lower cost of both of both exit and voice, particularly affective voice. So it's much easier to leave, in many cases, an online community than an offline community. And so there there are dynamics there that Hirschman didn't anticipate. Speech and to to have widely broadcasted speech, right, online. And so and so you see this through this, a preponderance of affect affective voice and, and exit options. And and, actually, what you see is very, very little, of the kind of of the kind of effective voice forms in which participants can actually directly have have binding effects in their in their, communities. And and the other, pattern that we notice is consolidated control. That that control is much more, potentially much more powerful, online, in in these online communities where, for instance, somebody has, kind of unchecked administrative power, over over a over a community as well as, you know, the kind of typical case of something like Mark Zuckerberg where you have, you know, one person controlling an entire company. This is kind of, mirrored in the way in which at the community level, control through these through algorithmic processes can is is much more highly consolidated than we tend to see in offline civic organizations where, you know, the the power dynamics are much more, you know, are are there are a lot more checks and balances. So so in the that that context, you know, we we attempt to articulate a bit about why this effective voice doesn't appear to be as as widespread as as it might be in other contexts. And and and highlight the possibility that there is need for more experimentation of the on this kind of voice.
So I I can take it from here. So we're we're observing something about technology. We're not saying, oh, well, the Internet made exit easier than voice or voice easier than exit. We're sort of taking this view, I think of it in terms of equip I heard about Photoshop back in the day that so what was the effect of Photoshop on design? Well, it made good design better and bad design worse. The technology just kinda makes things bigger. And so exit became easier and there's more of it on the Internet and we rely more on it. But the exact is true same is true of voice. The exact same is true of consolidated power, that all of these things become bigger. And what we want to do is we end up kind of challenging two assumptions at Hirschman. One is the association of exit with effectiveness and voice with effectiveness. We want to decouple those to look at effective voice. And then another thing we wanna question that the Internet has, like, sort of effectively bought into as well. And the Hirschman certainly bought into and a lot of people who have used this framework have bought into, is a very rigid line between the the leaders and the followers, the governed and the governors. That one opportunity of the Internet that we don't see exercised as much as the original rhetoric kind of imagined is a blurring of the line. There's more room for the people who are subject to some governance regime online to participate in it. And that's a and that is a I mean, we think it's a problem. That's our ideology going in. And and it's our ideology that makes that helps us sort of see the assumption baked in to exit to the exit voice framework, to Hirschman, and to a lot of stuff spinning out of it that that there's not a lot of room open to break that dichotomy. And so by challenging that dichotomy, dichotomy, by challenging the association of voice strictly with affective expression, we start to get an opening to make our argument. Baton.
Okay. Well, then we can turn to, you know, finally the the mechanisms that that we propose, which, again, should not be too surprising in this in this group. But we tried to highlight some areas of experimentation where effective voice, the effective could be done. And in some ways, these are just like exit and voice in general, kind of obvious. But we believe in this case, the obvious needs to be stated just because these things are so little practiced and so little presented. So the first category
This actually get yeah. The the need to state these things gets at something that Mako here Mako is here. Yeah. The sort of has an idea. Mako has been shopping around for the past couple years, which is that what's happened in online democracy since like Wikipedia, pretty much like where is all this experimentation and innovation? We're seeing amazing innovation and experimentation in exit. The internet gets off on exit. We've got crypto. We have all these platforms. We have eBay and all these markets. We have Facebook groups that you can leave. So exit has just flourished on the internet. But what what's happened to the democratic experiment that was the basis of a lot of the utopian dreaming about it in the 90s? Where's that gone? And so clearly, there's a need to say something, even though we all kind of want this and feel like it's there to be. There's clearly room for more energy and experimentation. So we do think it's worth just taking a position.
Yeah. And and and it, you know, comes with this sense that, you know, this this reminder this need to remind the people thinking about governance online that how much territory just has not been explored in significant ways. So so one type of mechanism that we discuss is mechanisms around authority and control. So just this can involve basic things like participants having say over over who holds administrative roles. So even something as basic as making a community's rules binding on administrators. Right now, you know, for instance, in a Facebook group, administrators can make, I think, 10 rules, But they are fully responsible for implementing them and are not really accountable to them directly because they hold all the power. And, you know, what if there was a, you know, there were mechanisms that held though those administrators themselves to their own rules other than kind of affective pressure from the from the group, as well as modes of selecting and removing those in authority. The second, box is mechanisms for collective action. So this is recognizing that a lot of the the user interaction dynamics in online communities are based on kind of one to one individual member to authority figure complaints or interactions. And what if there were more mechanisms for aggregating demands? One example that we consider here is, like, the obama era whitehouse.gov petition system that said, when a petition reaches a certain threshold of signatures, the White House has to at least respond to it. So so there's a a kind of threshold mechanism that requires those in authority, to respond to collective action as well as, recognizing the need for diverse forms of user input and and the ability to form blocks and parties and and interest groups within communities, which, you know, again, may not none none of these are you know, we're not claiming these are all universally necessary in all kinds of online communities, but these are all types of effective voice that are, you know, could be plausibly valuable in online communities and just are almost nonexistent. Finally I got it. Let me just add the the last one, then we'll get on this is mechanisms for community change. And here we're thinking about cases like like the recent conversion of Python from a benevolent dictatorship to a kind of elected board structure. What if there are certain triggers that, like, for instance, the dormancy thing in Reddit where if a admin is is not active for sixty days, others can apply to become the admin. If there are mechanisms that as certain thresholds get reached that require a participatory process of reshaping the, the fundamental structure of a community, so that the, you know, core kind of constitution layer, is, up for change through a member process. So those are the three boxes. Again, that's kind of mechanisms around accountability. Second, around collective action. And third, around community change.
And we had the fun with the the change one because, you know, when an institution, especially during SIES, but sometimes in response to outside events, institutions change. And those change points are rare kind of moments where people switch out of the mode of preferring incremental change and you have a rare consensus on a big institutional change. And those can be opportunities, to consolidate power in one person further and kind of create more distance between users and higher levels, or those can be opportunities to reassert a sort of deeper fundamental role for for users in the highest level of decision making. This this paper has been a real adventure for me, and Nathan's been really patient. So I'm I'm not used to there being a paper if I didn't do a big data analysis and if I'm not reporting the statistics and what those, you know, say about a theory or about some idea. And so to have a to be writing a paper that's just words and no p values, where the work is writing, that was really unfamiliar to me. But a good illustration of how it's work, which again is like an adventure for me, was in pinning down the some of the collective action side. So what's a like? Is a like effective or effective? What's a what's a a Kickstarter campaign? Is it like, how effective is that? What's a what's a what's a chain like, a change.org petition? Is that affective or effective? Like, did it did it do work? And where we ended up, I mean, I guess this is the inevitable academic cop out is there's a continuum, but we were able to con Nathan, just like very out of the blue, came with two criteria that we were able to use to really say, you know, like, retweet, you know, all these suggestion box at the store, a pull request, like GitHub pull request that we were able to say, what may where where's our line? And to the criteria we came up with or that he just, like, Athena ed out of his head were binding binding this so that this leads to this can, in some specified way, lead to lead to a choice that I know the decision maker is going to have to make, you know, on the basis of this. And so petitions. Whitehouse.gov satisfies that there's a threshold, you know, a 100,000 and the administrator administration promised to respond. Whereas the change.org doesn't have that binding aspect. And the other thing was transparency. So, not only does Facebook news feed not obliged to listen to your like in ranking stories that it promotes, It's also a black box. We have no idea how the like transfers transitions into a decision to promote something at some level on Noob's feed. And so those are our criteria. Transparency of the aggregation of many of these sort of expressions into into an institution scale action and some level of bindingness. And so when those hold a like button or or a flag of some abusive content or retweet is effective and when to the extent those things are absent, that mechanism is effective.
Great. So I I think we can leave it there. I would just say that, you know, I we're we undertook this kind of to I think in the hope of clearing space and opening further you know, opening the conversation for the kinds of mechanisms and discussions that we've been having in this group altogether and the kinds of projects that many of us are working on. And so in some ways, it feels like we're trying to spell out something that feels obvious to us in certain respects, but we've also been trying to be very precise about how to do that in order to convince others who might not, you know, see the value of the kinds of governance mechanisms that we're all interested in, you know, to to take this the need for them more seriously. So we'll leave it there and love to unless, Seth, if that sounds good, ready to open it up?
Yeah. Let's let's let's. We could just go in order of questions from top to bottom on the chat. And then when we hit bottom, open it up to questions that weren't written down. Are there questions or is it all comments?
Maybe see. Does anyone just wanna come in early? Maybe you've had a question in the chat. Yeah.
Well, I'm gonna just respond to okay. Maybe I should I should repeat my question before respond to Amy's answer to my question. So the question is, so so you're talking about affective versus effective voice and there are ways of converting between them as well as well, I think Seth mentioned just now like bindingness as a kind of like distinguish as a differentiator between affective and effective, but there are ways of converting between kind of the instances of active of affective determining turning them into more effective kinds of voice. So sometimes that occurs through, like, kinda like a community organizing process. Like, you can imagine Obama going to Chicago and kind of, like, getting sort of angry affect sort of angry affective, you you know, people like complaining or sort of expressing discontent and organizing that into a more effective local community. Right? But my question was like what are kind of the relevant ways in which this happens on the Internet? And I'm specifically reminded of, you know, mechanisms where as you've described several times where there's like a, you know, you you set some sort of protocol, there's a threshold, there's after enough likes or hearts, which we typically regard as affective, that converts into directly into like a a ban. Emmy's question to that is, like, how is that different from a vote? I think it's different from a vote because it's specifically it's like it's kinda like a it's a kinda like a dragneck for like, people will sort of, like, say, like, do likes and hearts or they'd, like, do dislikes and things like that as a way of expressing, like, their effective sort of sense of something. Not necessarily because they, like, don't want this thing to get banned, but a kind of like a controller or governor could look at that then and say, like, well, you know, if there's enough dislikes, then maybe I don't want that on my network and so use that signal. You use that affect as a signal to sort of say, actually, you know, people don't really like you, so you're costing me money or whatever it is. So, like, go away. Right? Of course, that can lead to kind of certain kinds of majoritarianism, but that that's why I would think of it as different from a vote. A vote is much more of an intentional, like, I'm explicitly participating in this kind of effective decision. Whereas an affect is more performative, I think.
Yeah. I think I I think I agree with what what you're saying. I also, Jenny has been raising some questions about the kind of line between affect and effects. Jenny, do you wanna to bring that in?
Oh, yeah. It it was a little bit rhetorical, but I thought it was a well, this is my first time hearing about exit voice and loyalty, and that was really interesting. But I was wondering if it naturally kind of settled into the roles that it has because there's something uncomfortable about affect in being able to actually enforce things like banning people and there's, like, majoritarian and, like, the general backlash of cancel culture right now. And I was also wondering how effective voice can be if it doesn't actually result in change, which I think you addressed with some of the bindingness and, like, relationships to, like, being able to actually implement change. But there something about that two by two is not, like, a totally even split.
Yeah. I think I think that I I'd love to hear you raise that in the chat. I the cancel culture question because, I I think there is a, you know, a power in, you know, that that affect can sometimes be very weak, but sometimes it can be like overly strong, right, where a few voices, end up overruling, you know, a much I hate to say it, but a kind of silent majority or something like that, where where, you know, affect can can compel a kind of effect that doesn't end up, occurring according to a kind of transparent process. And and that's a kind of case where we might recognize that relying too much on affective participation, might be damaging to something that we value. Like, for instance, due process. And, or having some kind of shared committed process that that, you know, in which we adjudicate conflict. And and that maybe when we rely on affect, we end up with a kind of vigilante justice. We also kind of, you know, we we start the essay kind of weirdly with with Martin Luther King talking about the riot as a voice of the unheard and explore some research that looks at, practices like trolling and, you know, other kind of potentially harmful online practices as being an outgrowth of actually not having effective voice, not having the ability to be heard resulting in kind of destructive forms of affect.
Yeah. This gets a another distinction we had to make. You don't want to take it uncritically when we say that voice is primarily affective. Look at it. Look at a protest down this down the street. Look at tearing down Confederate statues. So, we think of these as voice. They're certainly affective, and they're and they're quite, and they and they can be capable of institutional change, the kinda concrete institutional change we're calling effective. So how do we solve that? And we kind of ended up drawing a line between outside the system and inside the system change to some extent with the argument that, sure, lots of voice is effective, but it's but all the mechanisms for for transitioning voice into action, you know, so they so often tend to occur outside the system. You can look at the Reddit block out. You can look at, cancel culture. You can look at a lot of street protests. And so to the extent that we incorporate effectiveness intentionally in the form of mechanisms inside an organization, presumably, we can give a channel, that that, more systematically harnesses the energy or, you know, prevents outbursts or and makes people feel heard.
I've seen Divya has a question around effective and effective exit, which I know is yeah, which is something that Seth has brought up a lot in our conversations. And Divya, do you wanna say more?
Yeah. I was wondering and I I think it kind of jumps off of a lot of the discussion we've had in the past couple of minutes. But, you know, not to spectrumize everything, it does seem to me that there's also this, like, exit is also can be very affective or effective. And even in discussing cancel culture as we just were, I mean, to an extent, some of that, I think, is communication of affective exit or Jenny brought up these, like, delete certain apps, campaigns, or, you know, how do you how does that fit into this conversation? And is effective exit something that needs to be built in the way effective voice does, or is that sort of a separate piece?
Yeah.
A lot of the a lot of the previous sort of work with Hirschman has been taking this, like, two boxes and, like, adding a dimension and then trying to fill in all the boxes. And we kinda resisted that. There's certainly probably affective exit, but we just sort of wanted to dive into what we think is important. We wanted to turn a lot of this work into our rant about how great democracy is more or less. And so we just sort of focus. We didn't really like the the work wasn't so much to fill out the two by two. The the work was to make the case for bringing voice in more. It's funny. I mean, Hirschman's book, the way it's been interpreted as a sort of balance here is these two constructs, and they have this important dynamic, exit voice. But you actually read Hirschman and it's more just a polemic for voice. And and we're comfortable kind of running with that baton, presenting, you know, what might look like a two by two, but really kind of having our agenda.
I I wanted to sort of start by I noticed when when Seth was, talking a bit about not having the the the data that one thing that jumped out to me about what you're doing is that you're you're actually working on something that is, in some ways, almost more formal in the sense that if you continue to deconstruct the mechanisms at play and they're, like, the way they compose with each other, You're going back towards something that is, I I'll be at more more theoretical is reasonably formalizable in the context of, you know, a data structure that doesn't exist or that you wish that you could observe, but that you have sort of lived experience and sort of a phenomenal phenomenon understanding that tells you that something like, you know, if I could measure these things, I might see things like this. And, you know, that stands out to me in particular because I have similarly had concerns with the absence of voice in the decision making processes, and I know we've talked about in the past, but, one of the main ways that I see this being problematic on the Internet is the high frequency of action. So the the lack of frictions, the ease with which people exit things, or the even when you have effective means of interacting, everything is instantaneous. And so, you know, taking a slightly more theoretical, like, systems oriented lens, attend towards asking not, about whether you can or can't do something, but sometimes, like, how long it takes or how hard it is to do something actually material affecting the choice of mechanisms people use. So the kind of closing point there is that in the context of some of my work for our Internet related systems like the conviction, sort of voting slash, sort of, signaling systems. The the reason the low pass filters are there is precisely to give people effective voice that leaves room for effective voice because the current pilot show basically that people post proposals, they can't pass instantaneously, people are participating based on, you know, setting aside the the means through which we sort of provision the rights to participate. The community boundary definitions are are still things that need work because of the the tokenization aspects. Nonetheless, the communities are relatively well bounded, meaning that the pilot groups are participants who are really genuinely interested in exploring democratic forms. You see actually people writing up proposals for things, you see them going into dockets, you see them having movement towards passing where the voted participation is effective, but slowed. And the result is actually discourse on our discourse forums and in the discord channels around those proposals and you see actual shifts and some coalition building, around what should or should not pass or things get taken down or the the actual discourse aspect of that democratic process is, we'll say, nurturable. Not by no means, like, a fully solved problem, but just by reintroducing a slowness at the same time as bindingness, you leave space for both affective and effective voice.
That's nice. It solves the problem that exit mechanisms are kind of easier to encode in code than like, you know, deep conversations. And so you're almost making room for for voice not by acting on voice, but by slowly by using the power you do have to affect exit to slow exit down. That's I like that. That's clever. Niko, what's your two by two?
Well, so I mean, the the fir I think the the primary source alright. I'll back up. So I think that I think that you're reading of Hirschman and the conflation of, like, being able to express what you want and being able to change things is totally spot on. I think that the my primary sort of, like I mean, my my question for you is if if you've read or know of Kat Trego's book, The Conversational Firm, because my reading of it is that it's that it's also trying to build off of exit voice and loyalty, and it's basically making exactly the same, Like, the key takeaway from the book is exactly the distinction that you're drawing between sort of, like, unpacking I mean, the the those two senses of voice and arguing that that at least in in, like, the study of organizations that, like, people have typically complained at those kinds of things. That when they're talking about, like it's not about exit at all. It's just about voice. But when you talk about voice within an organization, like, the ability to, like, say how you think should think should be are completely conflated with the ability to, like, actually enact change in an organization. And her she's studying a firm that is, I don't know, like, a new kinda start up y kind of organization. And she's and it's sort of like a book for, I don't know, people that study, you know, more traditional sort of, like, large firms and is documenting the way that within this firm, people can like, employees of the firm can speak very openly. They can critique and raise major issues in ways that in traditional organizations they might not be able to do. And and her argument is really, like, based on Hirschman's sort of, like, distinction in that space. So, I mean, I I see sort of implicit in her work is the the two by two. I can probably, like it's like the it's the she's sort of arguing that, like, that the effective she's interested in what happens in this, like, the this spot, you know, down here, right, where you have the ability to express your interest but not actually really the ability to change anything. Because within the firm, it's still like a hierarchy managed firm. Right? There's still, like, a boss at the top who's gonna make the decision. What they've effectively been able to do is sort of open up participation to allow people to to to allow employees in the organization to tell the managers or the principals of the organization what they, like, you know, to be really honest, to view the financial information, to view lots of kinds of things, and to engage really sort of deeply. And so, I think that, like, like, one thing that I've always thought since I read the book that I was interested is, like, it's it's sort of, like, the other the the the other space. Like, what are the places in which people could, like, enact change, but maybe not, like, sort of express it in the same way? Is there some sort of, like, option for something like that? So that's sort of the the the it's just like a pointer and then, like, a and a thought. Because I think that even if your goal is not to I mean, I I I I'm I'm sort of with staff in the sense of, like, adding another dimension and then sort of exploring the the two by two by two by two as it sort of gets bigger and bigger is like a a sort of, like, a decreasingly sort of, like, useful game. But I do think that you haven't you're you you have even if you put x to the side, you are adding an additional dimension. And I think that understanding what's happening in that in the in that other spot, maybe it's not possible. I think that's, like, a a useful thing to think
about. Thank thank you so much for that. And, you know, I think you were the one who first got me on to the Turco book and and I need to revisit it. I I hadn't thought about it in the context of this work, but we'll definitely revisit that because if, you know, if, you know, the way you're describing it really does sound super helpful. And and
I think she also is engaging with Hirschman very explicitly. Right? And I know that other people that have picked up that book have also said, like, you know, Turco's new, like, approach to active voice and loyalty. And I think that that distinction is very similar to the one you're drawing.
Cool. Cool. Yeah. And and the point about the the expansiveness of the of the term, you know, I think just has to do with what our priorities are. And I think, you know, certainly in the course of of developing this paper, we've we've written a lot more than than we than we can fit in the context of the of the publications that we're aiming for. And we've had to make a choice between whether to make a kind of thoroughgoing, almost metaphysical theoretical claim about about, different sorts of exits and voice in a in a broad range to the more specific challenge of highlighting a particular form that is missing in the particular context of of these online communities that we're considering. And, you know, for better or worse, we've ended up going with the less metaphysical, more specific claim. But but there is certainly material there for making, you know, for making it a much broader claim.
Two resources that we were that we managed to get a lot out of, there's Sherry Arnstein has the ladder of participation, where she kind of creates a hierarchy of forms of effectiveness and sort of denigrates, like, suggestion boxes and near voting as, like, tokenism. And then, you know, what you wanna do is climb up the ladder, and the top rungs of the ladder are, like, deep participant involvement. And so one I I was proud of us for not fetishizing that too much and for kind of arguing for you should use all the rungs of the ladder kind of approach, arguing for a kind of continuity and sort of institutional diversity is what I would call it. Another resource we got out of that says tells the same kind of story you're describing Mako, but in a more antagonistic way is this woman, Gerwer I'm sorry, Gerkin. What's her first name? It's a exit voice and disloyalty is the paper. And what she's doing is is describing bureaucrats in, like, the deep dungeons of state buildings who, you know, they're not elected officials. They're, I guess, deep state. And they they care, and so they drag their and so they'll drag their feet on stuff they don't like, or they have a lot of the choice in how they implement policy. And she identifies this as a third axis sort of of agency that it's not voice because they're not complaining to superior. It's not exit because they're still in the building. But it is them taking an active role. So we treated that as some of our inspiration for exit and also some of our inspiration for breaking down the dichotomy between the leader and the led.
There's an Internet based equivalent of that I see occurring in the implementation of sort of platforms of smart contracts, whatever, where there's a policy that's written in essentially natural language and then its implementation is left to something significantly lower level. The code itself is the realization or the execution of the policy. And there's actually a lot of room for this, like, degrees of freedom in the implementation that can be exercised by people. And I see it in data science. I've I've seen it in in smart contract development. I had a relatively long discussion about it with my my my former boss and mentors, the CTO of Charter, which is the second largest ISP in The US right now. And we were talking a bit about algorithmic policy making and the extent to which as long as decisions that are made that affect a lot of people are executed via algorithms, as you actually have to be really careful about the gap between the policy making itself and the, the sort of execution of it and the degree of degrees of freedom that are left to the execution of it. So I think that's actually a really important insight and we have to look to the places where the dichotomy not only breaks down, but actually a lot of the agency sometimes resides. Like, the the decisions that are left are actually very sensitive for many people's outcomes.
Okay. We're coming along. So, we're down to five minutes. Maybe we could get a question from someone we haven't heard from as we wind down. Or from anyone?
I guess one access of exit online, and it was raised a little bit earlier with the idea that you could leave a community, but your words are still gonna be there forever. It's like you can't really opt out at some level. So because these organizations are nested, one organization's deep state set, as you put it, is another organization's working group where individuals may have not to have different types of voice or to leave that working group. But at another level, there's, like, another cash fall that's harder to exit. And so we're all Internet citizens. And so in that sense, it really has been awesome to hear about these ideas and resources related to what like, the letter of participation. Like, what does that base level of participation look like so that even if people are opted out of voluntarily or involuntarily from other online groups, what is the lowest possible standard, or what's the largest community that they can exist within?
Yeah. So you're getting it a couple of things. One is well, I guess so with the state something that in a state context kind of bureaucracy context, the Gherkin, interprets as disloyalty. This foot dragging is disloyal. In our framing context, that's really more of an expression of loyalty and commitment that you're putting time in to change this system in a way that's in line with your values. The other thing you're getting at is a little bit the blurry line between exit and voice so that we were able to find some literature on what's called sort of voice after exit or or exit is something that reinforces voice. So you complain, that's one thing. But you complain and leave, people give that a bigger message. And so generally, yeah, they're they're deeply intertwined. We're comfortable with that. That's great. So Nathan, I'm gonna let you take it from here.
Well, I think we we can just wrap up. I I really appreciate the, you know, all the questions and the and the feedback. You know, we're still very much working through this and and working on trying to hone and clarify what we're doing. But, what you all have brought up is really helpful in helping us to identify places that we need to make sure to patch. And and, you know, I hope finally that this that this all the energy and really fun conversations that we've been having over the last over the summer, you know, end up being of use and helping to draw more people's interest and attention to the to the kind of work we're we're doing collectively here. So thank you all so much for for being part of it. And, you know, it already this project already was an outgrowth of this seminar, but all the more so. And sorry about the about the about the last minute cancellation of what was supposed to be Probably a much better presentation for this session, but we've got that rescheduled for for a few weeks down the road. And we've got some really good stuff coming up. I believe Casey Fiesler is next. Yeah. That's right. So Casey Fiesler, my my colleague at CU Boulder, who is phenomenal. She's just, like, one of the most brilliant people and and one of the best people out there thinking about governance and online communities and practicing it in all sorts of ways from making comic books to to to helping to lead organizations and and much more. She's a lawyer as well. So definitely don't miss Casey next week. Thank you all.
Wait. Before we all leave, could I ask people to unmute for a second and give a round of applause for Nathan and Seth? In three, two, one. Thanks, Kevin. It's great presentation, guys.
Thank you all. Be well.
Welcome to the new