Institutional Grammar Metagov
Speaker 1: Alright. Today, we've got today for meta governance seminar, we've got Saba Siddiqui at Syracuse University and Christopher Fonts, professor of computer science in in Norway. Which which university, Chris, is that?
Top Keywords
- rules 0.023
- grammar 0.021
- institutional grammar 0.018
- institutional 0.016
- syntactic 0.009
- action 0.009
- chris 0.007
- constitutive 0.006
- statements 0.006
- well 0.005
- parameterizing 0.005
- basically 0.005
Transcript
Alright. Today, we've got today for meta governance seminar, we've got Saba Siddiqui at Syracuse University and Christopher Fonts, professor of computer science in in Norway. Which which university, Chris, is that?
NTNU, Norwegian University of Science and Technology, I believe. Very long. Yeah. So
Great. Great. And we'll be hearing today about their extensions to and Crawford Crawford and Ostrom's institutional grammar, what they're calling institutional grammar two point o, as I understand it, and, extension of the framework, in the direction of computational tractability. Saba and Chris.
Great. Thank you, Seth. We're so delighted to be here today to speak with you all. In today's presentation, I think what we're really interested in just doing is Shok is talking a little bit about what the institutional grammar is that Seth referred to you and what might be some relevance of this analytical approach, in the context of the various, studies that you all are interested in, in connection with what I see as topics of interest on the meta governance website. So we'll Chris and I will both be sharing some thoughts on this throughout the presentation. Let me go ahead and put this in slideshow mode. All right. As we get started,
I do wanna started, I
do wanna give you just a sense of who we are, and that will help you, certainly process and make sense of what it is we're doing and how we approach this work. So we are, an interdisciplinary team of scientists that are interested in understanding rules that are used in governance. And so I am a policy analyst, and my training is in public affairs. And Chris is a computer scientist, and so we're sort of thinking about this from different disciplinary perspectives, but in terms of common sorts of interests. Why we're here is, you know, because I think we have some something to add in helping you all potentially in the domains that you are interested in. So I took for this literally from the meta governance website, and I'm gonna offer my own interpretation of of what's included there. And, hopefully, my interpretation is somewhat accurate, although please please push back if if not. But it sounds to me that this group is convening scholars from different disciplines and different backgrounds that are interested in governance within the context of online communities where you see a distinctive value in understanding the development and evolution of rules rules, excuse me, that are leveraged in online governance. There's an appreciation for the ability of users to construct these rules, to organize with with one another, but also think about how to work together through collective action, organize social and political institutions that are relevant in these domains. You're interested in describing, supporting, and expanding this. But what really I think could be some linking points for us is that it sounds like your there's an embedded assumption, in the work that's being done here, that what happens, within these online communities, the way that users behave, the way that actors are interacting, so to speak, is enabled by features of the platforms, these online platforms, as well as feature and and these features of the platform also have some implication for the types of rules that don't govern their end. So this these last two bullet points are where there could be, from our perspective, some some use of the institutional grammar because, fundamentally, the analytical approach that we are working on developing is designed to describe, systematically, rigorously, reliably, and comprehensively the sets of rules, that are used to govern collective action. And the particular types of rules that we're interested in are both those that structure domains of interaction in the first place as well as structure behavior of, rules in these in these domains. And so the at a very high level, what is the institutional grammar? Well, it is an analytical approach, that's inspired in, linguistics that offers a way for describing the structure of institutions, including strategies, norms, and rules that are parameterizing features of domains in which multiple actors are interacting or are regulating actions of actors within these domains. So to connect this back to what I what I perceive is some interest of those of you involved in this meta governance project, we would think about the the, you know, the the the the point of, you know, understanding platforms in which individuals are interacting as linking to understanding of these rules that actually play a role in parameterizing arenas in which people interact. But then we are also interested in these rules that are specifically regulating with actors, and this approach allows us to distinctively evaluate both of these. And I'll talk about more more about that in just a moment. Our particular domain of interest, although we are an interdisciplinary team, we've been developing our our approach specifically for understanding collective action within policy public policy governed domains. So where we're speaking about policy is sort of, you know, formal, codified rules that governments have adopted to address particular kinds of collective action problems. And, you know, just reading in in the description of what this group is interested in, the meta governance group, it might be useful to think about the kinds of rules we're interested in as formal institutional structures that are somehow governed better collective action domains as well as regulating behavior within them. So I guess sitting at the core of the institutional grammar are institutional statements. So these are the institutional statements are the focal units of analysis that that one starts their assessment of rules with. And we basically think about rules and these institutional statements as akin to directives or individual kinds of rules that might govern collective action within a particular setting. And we think about those as being and taking one of two general functional forms. So we think about an institutional statement that can be regulated of kind, and this regulated statement is what regulates behavior. And we think of institutional statements as also potentially being of constitutive kind, and it is a constitutive statement that we, expect would constitute or parameterize features of the domains in which people are interacting. Now, this functional difference between, rules, becomes important to us not only because we see that rules, that we observe in our distinctive domain of public policy, but in other domains as well. Not only do we see different types of rules actually, right, doing different things within the system, if you will, right, we we find that, they tend to have different structural representations as well. So I'll turn now to talking about the linguistic aspect of institutional grammar. As I mentioned, this approach is, right, designed to get systematic, comprehensive representations of the structures of institutions drawing on this linguistic approaches. So what the institutional grammar fundamentally does is direct the institutional analyst or the the the analyst who's interested in understanding rules that govern behavior to get them to understand what are what are those individual rules and then guide them toward parsing those individual rules along different syntactic elements that have rule relevant meanings. And there are distinctive sets of syntactic components that that accord with different types of statements. So we assume, that regulated statements or, regulative, directives again, these are ones that are basically telling actors what they can, they can't do, within different, situations, under different constraints, and in in the presence of different types of incentives. They are basically assumed to be constitutive of these syntactic elements that you see in the left hand box. And so we have, you know, a regulated statement as being parsed potentially parsed into an attribute, which is an actor that carries out an action, an aim that delineates the specific action itself, context as representing the constraint on that action, an object representing the receiver of the action, a deontic that is a prescriptive operator that describes the how strongly an action is compelled and then or else which represents sanctions. And then on the other hand, we assume that constitutive statements are that are constitutive statements or constitutive rules more generally are comprised of some configuration of the syntactic components you see on the right hand side. So what we see represented syntactically and also semantically in these statements are something that's been constituted, which we call a constituted entity. We see some type of properties assigned to those constitutive entities with a constituting constitutive function linking these two, some, idea of constraints in which this constitution is happening, some level some indicator of prescriptiveness around this constitution, and then a sanction, for violation. So fundamentally, what we're doing with the institutional grammar is identifying what is the function that the rules play within collective action settings, and then further characterizing those along, those these syntactic components, per their functional properties. And so, ultimately, we have an opportunity to purse institutional statements or rules in syntactically as in order to ultimately infer what their meaning is as relevant to the domain being studied. Now I'm going to connect a little bit back to this other idea of platforms and thinking about regulation or behavior or interaction among actors and their behavior within certain platforms. Well, we we also have a conceptual platform that we rely on for thinking about, rules that govern behavior. We're assuming, based on the framework that we draw on, that any situation in which you have two or more individuals interacting, is is, called an action situation. And we further assume that any action situation is commonly governed by some configuration of the following of of the rules that are listed on this page, where the rules, are are are represented in the kinds of, right, in the institutional statement conception that we just reviewed. So we would assume that any any situation, in which you have two or more individuals interacting with each other is going to be governed by a set of boundary rules that identify who can participate in an action situation in the first place, position rules that identify who can occupy different positions, choice rules that give sense of their specific action sets, information rules that govern information flows, aggregation rules that govern decision making, collective decision making, and then and payoff rules that govern costs and benefits of taking particular outcomes, and finally, scope rules that identify intended goals or actions intended goals or outcomes. And the key thing here, again, is that we use this heuristic as a starting place to identify types of rules which can then be further categorized in terms of our specific kinds of regulated or constitutive rules, which then be further parsed along with syntactic components to get comprehensive understandings of the rules that are governing behavior in these collective action settings. And so I'll say now a bit before turning it over to Chris about how the institutional grammar two point o is is representative of our interdisciplinary interests. So one of the given, you know, my interest in policy analysis and Chris' interest in in computer science, we, engaged in some modification to the original institutional grammar to ensure that we are, developing an approach that allows us to understand rules in an ontologically consistent way, noting that we're we're relying on the syntactic representations of rules. But, also, we are developing an approach for classifying rules that is computationally tractable. And so we've built a framework in our revised specification of the institutional grammar that basically parses institutional statement information or rule information with varying levels of granularity under each syntactic component, recognizing that more granular representations at the syntactic component level may be necessary for engagement of computational approaches downstream, which Chris will talk more about in just a moment. So to kind of circle back to the top, so the institutional grammar is, again, this approach that we're using to describe norms and rules that are playing this parameterization role within the context of collective action situations that are governed by sets of rules, but then you also are playing this role of regulating actions. How do we see this I mean, how do we see the value of the institutional grammar? What do we assume that we use the institutional grammar for? Well, sometimes it's to code existing rules. So we see that there are rules that are governing a particular collective action setting. We just wanna get a better handle on what are the designs of those rules. So in that sense, the the institutional grammar can be used to basically just classify existing sets of rules. It can also be used in coding of emergent rules and Chris will talk a bit about how we engage the institutional grammar and agent based modeling where you have actually write rules in use so to speak develop among certain actors within a collective action setting. And the institutional grammar can be used as a ex post classification of pre emergent rules. But to speak to Seth's point at the beginning of this talk, there is possibility of using the institutional grammar in this in in the design of rules itself. So that particular use case that he was referencing, this can be accommodated within with with the institutional grammar as well. So So I'm gonna hand it over to Chris to talk a bit more about the computational aspects of the institutional grammar.
Thank you, Sawah. Let me reshare my screen for a second. I hope you all see my screen, at least an empty one for now. So let's bring up let's see. Right. Exactly. So the computational users of the institutional grammar, this is precisely where my perspective comes from. I hope everyone can hear me. But, so when looking at the grammar, I didn't see it so much from the policy, land specifically, but rather really what we you know, how we could how we could make use of it in a computational context and, you know, any mechanism that we could use to, operationalize it. And again, as Saba said right now, one of them is the coding of emergent rules. So, I'm coming from the back I have a background in computational social science, and there's all about modeling complexity and, emergent behavior primarily. However, in most instances, is this behavior not really overt. Right? It's not made explicit with that we're basically modeling institutions whenever we engage in those kind of modeling exercises. And they are the idea was or a central point is that, that's a core feature of the IG two point o, as well is to have a notion of nesting. Right? The statements, in in in its syntactic structure is Saba identified them, don't exist atomically, but are actually interlinked, related, and nest on each other, for example. And this is quite essential in order to afford any sort of computational representation, which I'm just motivating very briefly right now. So, looking at the idea of, having agents explain. So I'm thinking I'm talking here about a a simplified, agent based model that basically models and economic scenarios in which scenario and individuals collect information in the structural form as expressed, in the institutional grammar. Right? Having an actor, having aims, conditions, and so on attached, and then aggregate those in order to derive reasonably complex, in fact, very complex, well, institutional statement that nest on each other but comprehensively identify what the agents think, for example, about the institutions that are actually in use that exist in a system. And this is basically build at run time from the bottom up. And just without going into syntactic aspects right now, just highlighting here is the the the syntactic breakdown or representation of the statements in their structure, and you can quickly see how complexity emerges simply from that interaction. But the the the added benefit is here that it's represented in a form that is on the one hand, of course, computationally accessible, but on the other hand and that's probably more important, also human accessible. And that's generally not the case because we simply look at output oftentimes simply in terms of, you know, aggregate output. Right? So, this is just one of the application that would point towards immersions. Of course, that's merely a teaser. I completely appreciate it. But now having seen the community, I just want to also lose one or two words on the design of roots because that's also an avenue, that, we have engaged in in the past. So and this is really work that has been quite a while back and rather motivation, at least from my side, for the IT two point o, then, quintessential, latest work. But, most of you are actually thinking about designing institutions, I understand. At least that's my, my my gist here that I get from the introductory talk, or discussions that are overheard here. And that is really thinking about, you know, what is the mechanism that allows us to design institutions, especially in, you know, online communities, but more generally on the Internet, any sort of abstract interaction. And, of course, something that would be very comfortable you will be very comfortable with is the idea, of course, of distributor ledger technologies in in in specifically, of course, the ones of the second generation that allow the distributed execution of of code on the basis of so called smart contracts. Right? So it's basically little programs that are synchronized or executed across the distributed ledger across different machines. It's no longer sharing state, but also sharing execution state more specifically. And I'm I'm I would imagine that some of you, are very aware, of course, of the DIO concept, the the, distributed autonomous, organization in which the idea is basically that we no longer know only have actors that, of course, interact, but also that the governance is managed within this, you know, realm outside of any sort of particular intervention by a human actor in particular. So it's completely managed in the digital realm. And a colleague of mine and I, we were looking back at the situation. Okay. How can you represent institution in their purest form? And now the assumption is that the institutional grammar in the structure and form that Saba presented, of course, with extensions as usual in life, is is kind of the you know, an ideal case of actually constructing those contracts absent any further contextual knowledge and necessity to engage in it because it's purely represented in computational form, can we kind of map this into something that's akin to a smart contract and possibly directly or more or less directly executable, in in in this instance, for example, Ethereum or any sort of other, second generation, distributor ledger tech. And that's basically, I'd the the the the concepts that we have developed there, basically taking the fundamental basis of the institutional grammar with its components, of course, adjusting the syntax a bit to to accommodate human reliability while at the same time being accommodative to advanced features such as, you know, conditions representations, but fundamentally, it's nearly human readable. So the idea is here, you can now use this language, which is a kind of domain specific language, DSL, to express or decompose voting that, for example, characterize a particular institutional setting more generally. Here, for example, voting in an extremely simplistic form, of course, where we say that, you know, there are certain prerequisites of voting. First of all, of course, they need to be voters, but they need to be registered and then need to be both vote before a deadline, for example, and so on. They may only vote one once, and if a certain threshold of votes, for example, is aggregated more than 100, let's say, then this is not this is published, for for example, publicly. So the idea is really here, can we break down the system and then have this nesting constant concept where we combine all those individual rules into one rule system in a wider sense? And then something like this emerges from there. Basically, you generate a template smart contract that you, of course, in some instance, need to populate further, especially where executional semantics are concerned. But, fundamentally, you get the skeleton of this contract, and it is executable in such form, and it has certain advantages. First of all, the average developer, I guess, for smart contracts in the long term will not necessarily have a computational background, but will be more likely motivated from a possibly legal background or or or or akin to that. And thinking from a legal perspective and a computational perspective is, of course, not in 100100% congruent, especially if you need to accommodate specificities of a particular platform. Here, Ethereum, as an example. And staying on this level and simply representing it in, you know, pure rule form and then getting this output, you can, first of all, separate, of course, those levels of abstraction, but also automate the the avoidance of certain introduction of bugs as part of the coding, which which was one of the motivation of this paper. But fundamentally, the idea is here, you see the linkage between the rules and its component on the corresponding elements in Solidity, which is the smart contract language or one of the smart contract lenders supported by Ethereum such as structs, the voters, certain functions, modifiers, so those are the conditions effectively, and, activities that the system engages in. That's basically the idea there. Anyway, I don't wanna push this too far. I think I'm well over time, so I appreciate that I haven't been sanctioned harder so far, but it will happen in one or two minutes. So I I better come too close here. But the idea is basically, you see, you know, Samuel has has introduced the idea basically from the policy analysis perspective specifically and really motivated the development there. And I'm showing a bit more of the demand side where we actually saw the use and ability, you know, to use the institutional grammar in a real applied sense in in in very diverse disciplines. I give you a brief glimpse of computational social science and then focusing on the smart contract idea, and that's really showing the opportunities that arise from this by accommodating those refinements. Anyway, that's it from my side. I think now it's discussion time, so thank you very much for your attention.
Thank you, Chris. That was real whirlwind, but real red meat for this audience. Yeah. That was Could you could you post the link to, if you have a manuscript for that? I'm sure people are gonna consume it eagerly, and maybe someone would be, willing to launch that over to the, hashtag seminar channel on Slack. If you're, just joining the Slack, we usually keep a parallel dialogue going on in hashtag seminar, so be sure to join in on that. I'll be moderating. You can use the raise hand feature or I'm more likely to see it if you just put a little ping in chat, either your question or just say that you have a question in the chat. We're going to start. Josh had a question.
Yep. I can begin to respond to this question. I saw it in the chat.
Oh, Josh, why don't you go ahead and ask it? Or would you prefer I did? We have the audio record.
Can you can you hear me?
Yes.
Okay. If you can all hear me. My my Wi Fi is a little bit in and out. Yeah. So I've actually been attending solving Chris's kind of, I guess, what would you call it? The course on Institute for Grammar. And it's been really wonderful. Thank you for inviting me to that. And one of the questions I had that I guess I've been sitting on for a while is so I'm a computer scientist mathematician by training. And, you know, you think about parameterizations of things all the time. Right? And, you know, I was thinking that, you know, if insofar as, like, digital platforms are themselves action situations, that's kind of like something you mentioned, Tava. Mhmm. You know, I tend to think of software modules as forming on their own kind of, like, a parameterization of that platform. Right? Obviously, the the modules are set up in certain ways and configure it in certain ways. All the functions are sort of take parameters and sort of spit out parameters. And, of course, this parameterization is less much less friendly than much less interpretable than if I put it into the form of, you know, I think those statements. And I was just wondering what your thoughts, the both of you, about the relationship between, let's say, the kind of the the social science prioritization and the kind of, like, the natural prioritization that you see in a kind of in a software context.
Chris, may I may I go first and then Please. Pick up pick up on the more software specific questions,
I guess. So, Josh, this is this is a really excellent question. Thank you for posing that. I guess, one way to think about this is we in in our discussion of the institutional grammar, we're specifically focused on, capturing, systematically the way that rules are parameterizing, but we assume that there are other, factors as well that play a parameterizing role with respect to action situations. So the framework that we're leveraging, for example, suggests, from which we're deriving this action situation concept is institutional analysis and development framework. And within there, an action situation is is parameterized by a few different factors. One are, you know, material physical material conditions Some of another factor are is even characteristics of users in the first place and then you have rules that in addition to those other factors are parameterizing what's happening within action situations by way of parameterizing features of those situations. So the way that I interpret you interpret your question, correct me if I'm wrong, is that there are aspects of the architecture, borrowing the language from the website, that are going to pose certain types of constraints on what happens within action situations. But those are in some to some degree, separable from institutions or rules that also separately, guy parameterize action situations.
So I guess. I guess.
Yeah. Yeah. So in our case, right, so in other words, maybe even a a more a simplistic way to say it is that there are these non rule based factors that play a parameterizing role. And then there are rules that pay parameterizing roles. The institutional grammar specifically is designed to comprehensively capture, through this syntactic specification, the rules that pay parameter parameterizing functions. Does that does that kinda get at what you were saying, Josh?
Yeah. I guess, like,
the thing that
so there are okay. So there are regulative and cost constituted statements. And the constituent statements relate more to the architecture of the of the situation of the context. Mhmm. So insofar as code is a form of architecture and the laws codes kind of stuff, then you should be able to sort of interpret all the like, at least a large amount of platform architecture, like digital architecture into the form code. And that's exactly what sort of Christopher sort of brought up. Right?
Mhmm.
But I'm I'm just pointing out that, you know, there is a kind of of course, like, the the existing architecture has a certain kind of formal structure. Right? And, like, the more well engineered an architecture is, the more structured that thing is gonna look like. Right? So it's like I'm maybe this is more of like a data science problem. I'm not quite sure what is the right sort of framework for thinking about and relating these things, but it's sort of like, you know, like, when, you know, when I wanna code up, let's say, like, a a natural language document, often that natural language documents has certain kinds of structure that I can take advantage of when I'm sort of, like, trying to construct these institutional statements. Right? I'm just asking, like, okay. So there must be some sort of there is a natural structure in the design of this, you know, this digital architecture. And I'm kinda curious how well, you know, how in your minds these two relate to each
other. Yeah.
Perhaps if I'm can I offer a comment? Or so I'm I'm I'm just trying to interpret it. I think we can all come up from slightly different backgrounds and disciplines, so our interpretation may vary a bit. But the what what I'm taking away from from it, Joshua, please correct me at any time. So it's basically you're thinking about the abstraction between, let's say, module or application and platform underlying it possibly. Right? So, I mean, if you if I if I want to bring this narrative back, let's say, to to blockchain technology, you have Ethereum as a platform. Right? The Ethereum virtual machine, which is kind of a well defined, well engineered as you suggest, and and very, you know, explicit form of specification for for certain necessities and therefore represents somewhat a natural environment in which, let's say, smart contracts, which are also constituted and, you know, parametrization of the action situation in a sense that it it governs a certain, you know, it governs a certain set of behaviors, for example, and relationships between actors is embedded in. So I think it's more a matter of abstraction to look into, you know, what what level are you actually exploring at a particular point in time. Right? Do you intend to express a higher level action situation that embeds other action situations more specifically or, you know, this is kind of the the the the question you kind of probably need to to to ask yourself, where do you draw the the line? Right? So, I mean, I'm not sure if that responds in any way to what you think what you're suggesting, but I feel it's a matter of, you know, having embedded structures that sit within a wider, know, formal structure, which is akin to what we would call, let's say, the physical environment. Right? So in in the real world conception.
Mhmm. Are you feeling satisfied, Josh?
I don't know.
Yeah. Yeah. Thank you. Thank you.
Great. Great. Well, how about you pick it up, Z?
Cool. So the last thing that you said that really
sort
of resonated for me was this comment about the the the, like, level of abstraction. And and the question that I was sort of wanting to raise relates to, how you distinguish on the extent to which the framework that you're presenting distinguishes between, the the low level action situation or the sort of admissible actions to a particular agent within the rule set versus the sort of the state of the institution. And and in particular, in in my experience, it's very rare that if you define the rules at the agent level that you get sort of system scale behavior that you intended. And in in practice, especially with, you know, codable environments, you you can almost prefer to derive the action sets from the the set of conserved properties. So, like, declaring literally defining the the rules in terms of the, you know, higher scale scale properties you want to preserve and use rules that you know or would expect at you know, whether analytically or via other formal methods would actually preserve those properties. And so in the context of your work, I mean, we have, like, a we have a way to clearly distinguish between the things that are, you know, enforced at the individual sort of action level versus the things that we want to hold for the institution, you know, as a whole.
Chris, do you wanna
Yeah. Yeah. I mean, the the only thing I can think about is, again, the abstraction idea, of course. But, I mean, doesn't it fundamentally link to the ID principles of, you know, having an operational level, the collective choice level, and a constitutive levels that basically reflect this, and you will need to define and decide what is the boundary, right, in in which you're actually modeling to which you apply those rules. And, short answer to your question, of course, you can apply those rules on any level, but you need to be specific which level you're thinking about when you'll be employing and applying them. Right? So and then you can, you know, increase the abstraction level suggesting, okay. Let's build a constitute well, the wider environment in in a softer sense or in a a constitutive sense in in which sort of any decision making can occur, but also in which, more narrowly, any sort of or on a micro level, any action can occur. So I think this is really compatible and congruent with the ID framework in in my reading, but perhaps I'm missing an important point
that you said.
The thing
that I'm trying to tease at is establishing through the fact that these are computational relations between the scales. So it's one thing to pick a scale of abstraction level to define these things at. It's another to be able to make strong claims about the relationship between the model at two different scales. So we have the sort of you know, we'll call it the what I would normally think of as the system level scale. I I'm a systems engineer by training, like, formally. So, like, I have a higher upscale where I have goals, like, actual, you know, goals for humans, and I have a lower level, scale where, you know, the individual events are occurring, and maybe I only really can enforce properties at the lowest level, the rules about what one can do under what circumstances. And then some sequence of events is, like, occur according to the decisions of those individuals. But if what I care about is really the, you know, the performance or the properties of the aggregate system, then I'm I'm sort of asking about how you see go about going about, like, managing the relationship between the model at the sort of actor scale and at the institution scale. And that as it happens, that's, like, a main focus a lot of my work, and I really like what you guys are presenting. And I'm thinking about how to integrate it, but this is the key topic for me is, like, the relationship between the scale of the model as an individual decision maker versus the the system or the institution.
Let's see. Have we can we go ahead and chime in, actually?
Yeah. Thanks. So so, Michael, you you are basically if I understood correctly, I mean, you were asking about the reflectivity. Right? And that's kind of for what's was my question too. Like, and Christopher, I wasn't quite sure. Did you mention it? But do you see the the grammar? Is it reflective? Like, can the grammar, can the language talk about itself? Or is is that the direction that you're going? Because that that would be the It's so what is needed in the system.
That that that that's right. So I mean, the the the the idea in a sense that you have a recursive representation of the grammar itself. I I don't think we we haven't implemented this in any form, but it's would be principally conceivable, but that's not the intent of the institutional grammar. However, the the more explicit representation that we have right now affords a set of logical transformations that you weren't, you know, previously able to easily perform, I guess. But it's not reflexive in a sense that the the the the grammar can be defined in terms of its own components if if you think about this. No. So this is clearly a boundary where, basically, that's the that's the design limitation. That's where the the the the the environment, the abstraction level stops effectively in terms of devising the the the language, whatever. Yeah. So the the institutional statement, I guess. Yeah.
Yeah. Anyways, I really really understand. I'm I I got the paper. I'm I'm really happy to read. We we are working also on on some something quite similar in a sense, but we are after the reflectivity also. But
Yep. This is great. And when when you talk about reflectivity, just for for for for following up, if I may, is it are you thinking about agents or, sorry, entities creating their own, let's say, smart contracts, or are you thinking about refractivity really of the language itself? Right? Defining the language itself. Alright. Okay. Cool. Yep. It's a very nice one. So yeah. Sounds good. By by the way, the yeah. Sorry.
Sorry. I I was just gonna ask my question if that's okay. So, like, we've heard a lot about how institutional grammar can be used for, like, policy analysis and how it can be used on all these platforms like blockchain and stuff, but I'm more interested in its use case for, like, in person communities. Like, smaller communities are interested to transitioning to having an online presence. Where do you see, like, the value added of using institutional grammar for helping online communities articulate their value structures when they try to move towards encoding those value structures on an online platform? And, like, do you have any, like, ideas for good starting points of how to do that?
Chris, may I may I offer a reaction first?
Or, well, I think part of it I mean, so there could be some, balance in terms of using the grammar to recognize existing exist rules and use, so to speak, as a basis for thinking about designing rules moving forward, maybe formal rules insofar as you might think about the rules and use that develop over time naturally or have evolved within particular communities already in some way are reflecting valuations of some kind. And then so so before you can, you know, engage in the prescriptive, you may just look at the normative, representations of what rules are. Does that make sense, Samantha? Yeah. That
that makes sense. Thank you.
Mhmm. Mhmm. Sofia?
I think my question was somehow contemplated by Samantha as well because we are in a phase that we are creating rules for our community and seeing what are the emerging rules to start iterating with them. And I wonder how how does the language can, come from the community? And if it makes sense to have an institutional language, if if that doesn't relate with the people that are that are using the language to to put the rules in the prac in the cultural practices that we have.
Mhmm.
And I guess my question is, do you see rules emerging from the community? Do you see a process to create this language within the community from these rules that are emerging from the community?
Can try this for a follow-up? I don't know. I'm just this is wondering. It was mostly clarification question. So you're thinking about I mean, I'm just trying to reflect on this. Both Samantha and Livia, you mentioned the idea of values. Right? So and, the question that I'm challenged a bit here is that that is, of course, the most fundamental, you know, element that make up communities and and our mindset, I guess, in terms of social individuals. But, do you see the, values to come first and being expressed in rules, or do you see an emergence of values within given communities as expressed in the rules? So, basically, is the is the value of the outcome or the input of the process? That's the the aspect I'm missing. Or am I missing your point completely, which may well be possible? Right. I mean please.
No. I think in in in in what I'm trying to express, for example, our community has values that we came up together with. So there is a set of values, but because it's a decentralized community, it seems quite challenging to identify all the rules. But we know for sure that the rules are there, but maybe we haven't found this common language to identify the rules that are already emerging. And if you if you have an idea of a process to to do that in a collaborative way.
Just as immediate spontaneous response. I mean, if even that's how how we would do it in complex social systems modeling, if we're looking at the institutions in use, we look at the violations. Right? Because usually, violations are indicative of, you know, transgressions, and that is probably a good good point of, stepping back and saying, okay. What was what actually happening here? And, you know, what fundamental, value expectation or otherwise has not been satisfied by the, you know, a transgressing individual and then try to formulate a rule out of this one. So as a kind of, agile process in the wider sense that responds to those violations in in order to kind of shape then this root system more comprehensively. Anyway, that's just a spontaneous response. So not so much, but positive perspective there of in constructing the roots that are comprehensively, but actually looking at the violation and, you know, thereby defining further values such as, the the, liberalism, basically, that's embedded in your system as well. How violent for ex sorry. How, how how flexible are you to, for example, allow rule transgression in the first place. So it's yet another value that emerges just just by the observation of compliance in the first place. Anyway, just some thoughts. Very very very interesting point you're raising, Olivia. Thank you very much.
If I may offer a response as well because I I interpreted this in a slightly different vein. And so to just compliment what Chris is saying hopefully in some way. I mean, of Samantha and Olivia, I hear you both sort of getting us to to think about what are relationships also between rules and use and rules and form. And so one thing I would say is, you know, the institutional grammar, we're using it in the use case of policy analysis. But, actually, it is generalizable to rules and use and rules and form in its original conception. So right? So even if, right, you have communities that have exhibited certain behavioral patterns preferences, the idea would be that you can still represent these the rules and use that are represented in these using the grammar. And so it could be, right, a characterization through some sort of method of the rules and use, Right? The along in in syntactic terms, in the grammar terms. But then you can sort of then use that as a jumping off point to say what might be embedded in formalized institutions or rules and form based on what is represented through the grammar in rules and use. In that sense, I mean, because, Olivia, you said you're very interested in the Ostrom work around this. Origin I mean, this is this is the notion, to capture rules and use as well.
Mhmm. So, I want to on the side of ending before the hour instead of after. It feels a little gutsy, but maybe we'll be able to address Glenn's question. Glenn, could you ask your question?
Yes. Thank you. I was wondering if you've been if if I were a member of an institution designed in this way, have you had any experience of communicating it to a sort of an average user of how they can participate, how they can what actions are are, legal. So so communication of the design of the system to people who will participate and who aren't aren't programmers or or institutional designers. Yeah.
Yeah. Yeah. Mhmm.
Thank you.
Chris, may may I offer a response? Or please, if
you have if you have a
response, please go ahead.
No. No. No. No. I would argue from a software perspective, so I don't necessarily quite follow in terms of the community conception that's addressed here. Please go forward.
Well, I mean, Glenn, I can tell you how I I've done it when using the grammar and policy research. So, you know, like, oftentimes, what I'm doing is I'm taking policies that govern particular domains, analyzing with the grammar. And then the idea is basically communicating back to, like, what are the patterns that are observed in the language that might be relevant for understanding how policies are being implemented, whether they're working. And some some of that involves actually communicating back to stakeholders, practitioners, policy designers, regulated communities about the rules of these things, the designs of these rules. And I would say, right, it's drawing out, from the syntactic patterns what seem most relevant. So if there's a compliance problem, it's it's highlighting to them what information we see represented in the or else fields, that is the incentive fields. And and if it's conveying, like, some measure of complexity to stakeholders, it's trying to identify, well, what are the what is the what are all the activities that are represented in these policies as captured with these aim fields or these action fields? So it's basically, right, summarizing information that accords with different syntax components in a way that would be, right, suited to whatever is most interesting in a a particular governance domain. Does that kinda get your
Yeah. I just to behind my question is what I've hoped
Well, I'm I'm I'm afraid I'm gonna have to restrict any follow ups to the Slack, but I'll make sure that Chris is having an invitation to the Slack. So we so to close out for the day, we have a little tradition I'm quite proud of where we all unmute so to applaud together. So I'll count down to three for us to unmute. And thank, Chris and Saba for sharing, the extensions, the ID two point o that they did today. On 3123.
Thank you. Thank you.
Thanks so much. I'm looking forward to continuing this conversation on the Slack. Chris and Savo, I'll make sure that you get the invite.
That'd be great.
That'd
be great. Thank you very much. Thanks, guys. It's really great.
Yeah. Thank you so much.
It's a interesting community. Definitely. See you guys.