Constitution Suggestions

[Ticojohnny]

Now that GovGen is live, it's a good time to start creating a list of needs for the Constitution and focusing discussions on additions that should be considered.

For reference, the Constitution.

There already has been lots of discussion in the issues already such as:

• Limiting AI
• Governance and Community Pool Discussions
• Fixing Power Dynamics and comments here
• Establishing and monitoring conditions for providing ICS
• Enforcing the right to soft fork
• Improving Delegators and chain Safety
• List of Miscellaneous Suggestions
• Conversation on removing validator voting

There are likely to be many more, so let's use to this space to hash out and consolidate a proper list of suggestions.

[Stuyk]

I'm going to toss my hat into the ring and throw it in there for future proofing atom one.

I have a general suggestion that we should create a general procedure in the case of a quantum emergency where quantum computation has exceeded or broken cryptographic algorithms we currently use.

I think it would be wise to have a plan of defense similar to how Ethereum is going to be handling it.

While they are trying to focus on some form of quantum resistance, I think we should have a plan for that as well.

A. A validator creates a proposal to identify a timeline when the attack happened, and with relevant proof.
B. The chain and all of its funds have a snapshot taken before the attack.
C. If binaries have not been created with quantum resistance in mind, it becomes an all hands on deck scenario to build a new iteration of the protocol to continue the chain with quantum resistance introduced.

I'm sure we could get some more involvement revolving around this, but I'd at least like to throw this idea out there.

[cosmonought]

I have some general remarks and then some specific feedback on the draft currently here on GitHub (I've seen another draft circulating on Google Docs, but I'm not sure which is most current and was directed to post here on GH), so this post is coming in two parts with some footnotes to provide scholarly direction where it may be of interest.

For reference: my Discord/Telgram name is "Lil D," and I am a contributor to Neta DAO where I led the effort on legal incorporation and organized and developed the DAO's Constitution. I also organized and contributed to the development of the Charter for the Juno Network, providing research and editorial support.

General Remarks

1. Constituent vs Sovereign Power: A Constitution instantiates a power which is arguably different from sovereignty. Sovereign power, e.g., the divine right of kings, is the power of one over many; by contrast, constituent power (the term coined by the US Founding Fathers) denotes the power of many to become One.[1] As far as I know, the widespread marketing claim of "self-sovereignty" in Web3 has no specific intellectual/political lineage, so it seems to float a vague promise that might be paraphrased as: "all of us will be our own kings," without working through why a reversion (albeit in the form of a hyperbolic exaggeration) to a feudal model of power would improve the world.[2] This is just to say: the recognition that a Proof of Stake network's model of power is constituent rather than sovereign, and therefore needs a Constitution to properly honor and 'live up' to its reality, is a major step in continuing the revolutions of modernity that have been blocked and stalled under the occlusive weight of finance capitalism.

2. Second-order Rules: A Constitution is unlike regular legislation. Normal legislation creates what legal scholars refer to as "first-order rules," or rules that tell people what to do. By contrast, a Constitution is a system of "second-order rules," or rules about rules. In other words, a Constitution provides something like the rules of rule-making, so it expresses the values a community holds about rule-making and rule-following.

3. Entrenchment and Amendment: The two axes of Constitutional design are entrenchment and amendment.[3] Entrenchment is the means by which certain rules or clauses are made more difficult to change, and amendment is a specified process by which change is procured. Note that legislation can normally be overturned by other legislation, but that changing a Constitution usually requires "deeper" and more significant effort: for example, to amend the Constitution in the US, a 2/3rd vote of both houses of Congress, or a 2/3rd vote of all states to convene a constitutional convention followed by a 3/4th vote of all states to ratify). In the US example, the entrenchment of the Constitution and the Constitution's amendment process are basically one and the same, but a Web3 constitution does not necessarily have to follow this model: provisions can be entrenched at different levels simply by varying vote parameters, while the amendment process can become plural and tailored to the various areas where it is to be applied, etc.

4. The Five Criteria of Good Rules: Institutions—be they states, companies, or other organizations—do not, or ideally should not, speak like people. Rather, they speak in what has been called "ADICO syntax," a grammar for rules that enhances rules' clarity and implementation.[4] ADICO syntax says that a rule is well-articulated—i.e., that an institution 'speaks well'—when it address five criteria:

Attribution: to whom does the rule apply?
Deontic: does the rule permit, compel, or forbid an action (i.e., is an action optional, obligatory, or verboten)?
aIm: to whom or what is the rule targeting (where is it aimed)?
Conditions: when and where is the rule applied?
Or else: what are the "or else" consequences that follow from breaking the rule?

ADICO is useful when crafting both Constitutional clauses as well as normal legislation.

5. Conflict and Interpretation: Because a Constitution is a linguistic entity (even when it is 'encoded' in software), its meaning and interpretation are liable to change as communities and their linguistic practices change. Ideally, these changes are gradual and/or amenable enough that triggering the Amendment process will allow a community to update its Constitution as appropriate. But what happens when those changes are rapid or are not amenable to all (or at least to a clear majority)? and what happens when conflict emerges around the interpretation of some legislation and its constitutionality? It is important to identify a conflict-resolution process for interpretive disputes, and it is important for that process to have a sole final authority (even if there are many intermediate authorities on the way up, such as found in the US judicial system).[5]

Notes

1. For more on the distinction of sovereign vs constituent power, see Toni Negri's Insurgencies: Constituent Power and the Modern State, especially the first chapter. To be sure, not all thinkers agree with Negri, especially as concerns his complete dissociation of constituent and sovereign power, but I believe the heuristic value of the distinction makes it a useful and worthwhile contribution to political theory, even if historical polities have never reached—and indeed cannot reach—the state of "permanent revolution" that he identifies as pure constituent power.)

2. Curiously, the concept of "self-sovereignty" does appear in the philosophical work of Georges Bataille, who uses the term to designate the human subject's capacity for self-shattering, a power to destitute and reconstruct the ego. But this usage seems remote from Web3 claims.

3. For more information on Constitutions as second-order rules, as well as entrenchment and amendment,, see Eric Alston, "Constitutions and Blockchains," https://www.thecgo.org/research/constitutions-and-blockchains-competitive-governance-of-fundamental-rule-sets/

4. For more information on ADICO, see Cristy Watkins and and Lynne Westphal, "Institutions Don't Talk Like People," Policy Studies Journal Vol. 0 no. 0 (2015), https://www.fs.usda.gov/nrs/pubs/jrnl/2015/nrs_2015_watkins_002.pdf

5. Again with the example of the US: James Madison foresaw such issues arising (and indeed they did almost immediately) and withheld his support at the Constitutional Convention until his proposal for a sole final authority over Constitutional interpretation was included, i.e., the Supreme Court. A contemporary example of Rio de Janeiro, which has a Constitution but no final authority for its interpretation, has been politically riven in part by an interminable back-and-forth between executive and judicial interpretations, especially as concerns property rights and land usage. See Eric Alston, Tomas Nonnenmacher, Bernardo Mueller, and Lee Alston, Institutional and Organizational Analysis: Concepts and Applications, especially the fifth chapter.

Specific Feedback on ATOM One Constitution (as found here, though I don't know if it is entirely up to date)

Article 1.A All subsequent governance proposals must align with the provisions of the Constitution, and proponents of each proposal, along with all active governance voters, are required to ensure consistency between such proposals and the Constitution.

The "or else" enforcement mechanism here is unspecified or very weak (see General Remark 4 above, on ADICO syntax). Suppose that the community votes for something which is broadly popular but not consistent with the Constitution—which will prevail, and how?

Article 1.B Other sovereign zones that are completely or partially secured by AtomOne Hub ICS, by definition, have their own governance mechanism. The AtomOne Hub principally enables and follows the will of the governance of such sovereign zones regarding the pegged tokens originating from said zones, except in well-defined exceptional circumstances involving bugs, theft, or harm to the Cosmonauts of the Cosmos ecosystem.

The other two paragraphs of A1.B should probably be moved to the Definitions section. This substantive excerpted paragraph, however, is vague, and vagueness will lead to interpretive dispute, especially as concerns 1) the Hub's (or its validators', or whoever/whatever has voting power's) ability to accurately know the "will" of completely or partially secured zones, and 2) the "well-defined exceptional circumstances" that are not here defined—"bugs, theft, or harm" are all subjectively qualified. Suppose e.g., that a secured zone wants its token to do XYZ, but that some validators are concerned that XYZ is now illegal in the jurisdiction where they operate: how are they expected to vote?

In general A1.B seems to be concerned with explaining how the Hub relates to and behaves toward its secured 'spokes' on the ecosystem wheel. I am wondering if it is possible to completely separate the Hub's voting rights over secured zones, and simply letting security be security...then in the example case above, the question is not one of enabling or implementing the will of a particular zone, but rather the limits at which alignment breaks down or becomes unfeasible. It might be useful to require such zones and the Hub to have 'treaties' or similar agreements that anatomize and flesh out these details.

A1.C

This section could be included in the Preamble.

A1.D

This section can be merged with A1.E.

A1.E The Liberty and Property of all Cosmonauts engaging in the AtomOne Hub is hereby guaranteed. Any restriction to the Liberty and Property of Citizens on the AtomOne Hub can be done only through the AtomOne Hub's governance.

The language here is quite romantic but for that reason the section is difficult to understand how it is implemented, because it is so vague. What is meant by "Liberty and Property"? These are Big Ideas that can be construed and understood in extremely divergent ways, causing froth and discontent later on. As written, the text strains in a libertarian direction in its first breath, and then recoils to a (in the classical, not the modern pejorative) liberal position in its second breath, without clarifying which is supposed to hold.

It seems to be a confused expression of Rousseau's position on private property, where he wrote that "the right of each individual to his or her own private property is subordinated to the right the community has over everything." Or of Thomas Jefferson who wrote: "It is agreed by those who have seriously considered the subject [...] [that] whatever, whether fixed or movable, belongs to all men equally and in common[....] Stable ownership is a gift of social law." German philosophers of this period came to similar conclusions about private property, with Johann Fichte writing: "all property rights are grounded in the contract of all with all, which states: 'We are all entitled to keep this, on the condition that we let you have what is yours'." Hegel's gloss is more diremptive: "the right of distress overrides property rights."

At least one thing each of these thinkers has in mind is something like the problem of seemingly abandoned, or unkept, property, which abandoned property can "distress" a community not only by failing to contribute to it but even by actively creating nuisances and difficulties. Different jurisdictions have different rules and processes for the reclamation of such property—some indeed impose very high bars for reclamation—but virtually all jurisdictions have such rules and processes.

If a blockchain is going to survive beyond "the now" (say, the next 10-20 years), then it will need means to seriously grapple with problems stemming from abandoned or long-dormant tokens/wallets, which will inevitably arise as token holders lose interest in the project but don't formally leave, or forget/misplace their seed phrase, or even die. This problem may be more acute for abandoned staked tokens, and not as serious for abandoned unstaked tokens; it may be attenuated based on whether the supply is uncapped or capped; etc. Still, I think it would be wise to specify at minimum some conditions or criteria under which wallets can be unstaked from the network.

A2.A

While this appears to be another definition, it may make sense to add a clause to the effect: "and any additional functions/features/modules [or whatever] as voted on by the Hub." I am not a developer so I will leave the language to someone else, but this addition will make it so that what the Hub is can grow and the Hub Constitution can still apply.

A2.B UX interfaces that present the results of voting on governance proposals should also display the content of the memo field of each voting transaction such that the reason for the vote can be seen.

This article largely focuses on rules for governance proposals (the rules of rule-making), but it includes the excerpted section above that seems to impose rules on entities outside the Hub itself—I agree with the intention of it, but since those UX entities aren't necessarily bound by the Constitution, does it make sense to include this here? Perhaps this should either (a) be a piece of standard legislation governing the kinds of UX that the Hub will (financially or otherwise) support, or (b) converted to a rule that the Hub should only support governance UX that attains to such-and-such criteria...

A2.C

This seems unnecessary to specify, and includes a vague restriction that airdrops can slash "according to well defined principles"—but what are those? And moreover, ADICO: in the case that some airdrop does not adhere to well-defined principles for slashing and therefore is unconstitutional, what is the "or else" mechanism by which A2.C gets enforced?

A2.D

This section seems underdeveloped. What is the difference in a Treasury DAO and some other kind of DAO (and are other kinds of administrative DAO allowed)? The Treasury DAO has a specified Executive Board at the top level, but no other levels are mentioned or identified. An Oversight Committee "composed of any number of Cosmonauts" is mentioned, which seems odd. Moreover, the Treasury DAO requires "public and known real personal identities" of its members, but how will these identities be verified and who will have access to the verification—any Cosmonaut, or just the Executive Board, or...? Finally, there is no specification for how any Cosmonaut joins or is appointed to any of these entities, nor of how long they may serve. A rough conflict of interest policy is gestured to, but should probably be extensively articulated.

As a community we will need to thoroughly revise this section, and perhaps add a section on the use of DAOs and their membership rules more generally.

Part 3

The tokenomic rules given here are fine but feel incomplete.

The Common Pool, for example, should probably have some additional rules around it: for example, 1) Common Pool tokens should never be staked to the network because they have no rightful owner. So using CP tokens, e.g., in liquid staking protocols should be forbidden because these tokens distort the constituent power of bonafide stakers and network participants. Staking CP tokens opens a regressive problem where determining how the staked CP tokens should vote requires that a vote must first be held on how those tokens vote...2) Common Pool tokens should probably never be used to provide liquidity on DEXes, because it is legally problematic. And so on.

This would benefit from more discussion and wargaming out issues and concerns.

A4.B No zero-knowledge proof systems may be adopted on the AtomOne Hub even if they are composed of the approved primitives.

I'm curious about why.

A4.D

Unclear language. "Two-thirds supermajority of AtomOne Hub governance"—is the intention for the quorum to be a standard vote, but a higher (two-thirds) vote approval rate? Or does amendment require a turnout of two-thirds of all voters, i.e., a two-thirds quorum + a two-thirds approval? I think it would be wise to opt for the latter.

[cosmonought]

Some additional thoughts re: Jae's Comments and select GitHub posts:

Comments from Jae: The Economic Model + Staking vs Money

Jae writes that the original economic model for the Hub was not for ATOM to be a monetary primitive but to accrue "token transfer fees" in the mold of VISA shares. Staking vs Money repeats this argument but amplifies it with the security concerns of pursuing ATOM-as-money through parameter manipulation.

I posted the beginnings of an essay on ATOM as governance vs ATOM as money on the Cosmos Hub forum here and elicited a range of responses. So that you don't have to read that, I will say upfront that I am not entirely persuaded by either side: I believe that blockchain does bear an intrinsic and important relation to the question of moneyness, but that it also (or rather, by that account) poses a significant challenge to politics in the grand sense of nation-building, to the degree that these politics have become neoliberal, i.e., subsumed to the market.

I will try to TL;DR some of the research that I have conducted on moneyness through the Neta DAO Academy seminar Coining Reason. Section 1 is entirely optional.

1. Theories of money (optional)

Monetary theories are divided into camps, each tending to privilege one of the four functions of money (standard of deferred payment, medium of exchange, unit of account, and store of value) as THE function of money, with the others flowing as subsidiaries from the main function.

Speaking roughly and broadly: classical and neoclassical theories privilege the medium of exchange, Keynesians privilege the unit of account, Austrians privilege the store of value, and smaller but intellectually important (if relatively unsuccessful in practice) factions (e.g., proponents of mutual credit) privilege the standard of deferred payment. These tend to crudely break out into theories of commodity money vs theories of credit money.

Each of the four prevailing accounts also seems to have some drawbacks:

• The neo/classical positions are based on conjectural (i.e., false) histories of economic development that tend to smuggle particular ideologies into our thinking, naturalizing, e.g., 'competition' as a primordial fact of human existence;
• the Keynesians, buttressed by considerably more historical data, have more sophisticated accounts of money's origin and development, but their talent for descriptive accuracy seems to limit or blinker their boldness in theorizing or entertaining alternatives and their willingness to question whether their account is worth fighting for;
• the Austrian attachment to scarcity as a guarantor of value is deceptive in its mathematical persuasiveness, since value still has to be denominated in a unit of account;
• and mutual credit systems (LETS, etc) are intellectually interesting but face serious practical challenges, namely, that mutual credit broadly construed requires a God's eye view of the economy...but this is just what the "invisible hand" means is impossible: there is no vantage from which it is possible to totally comprehend markets, every such position is partial and limited, and the partiality and limitations become stultifying or fatal when elevated to a godshead.

One 'neat' resolution to this dogfight that Coining Reason has examined is the economic theorist Tony Lawson's account of "social positioning," i.e., money is defined by an act through which a society "positions" a given thing as money. Lawson thus rebuffs the neo/classical and Austrian fantasies of a spontaneous, emergent, or natural currency; however, at the same time, he reclaims the function of the store of value as the principal motor of money.

The question you may be asking is: What "value" does socially positioned money store? Lawson's answer: the value of so-called one-sided payments, i.e., taxes, penalties, fees, etc., often but not necessarily as demanded by the state. These one-sided payments lay the groundwork for both a unit of account and a store of value. What now enters the picture of any money's sustainability is the legitimacy (or perhaps the power of compulsion, the monopoly on violence) of the government. Lawson's argument chimes with the work of Harvard monetary theorist Christine Desan, who argues that money is a "constitutional phenomenon." For her part, Desan wavers between a Keynesian model, in which money is essentially decreed by a sovereign, and a notion of "constituent power" akin to Lawson's social positioning theory. This oscillation is understandable given that many real and historical moneys are indeed simply compelled by the state/sovereign; but this compulsion is clearly neither adequate nor sufficient without further buy-in to the legitimacy of the state.

2. Blockchain and money

Blockchains potentially sever the need for money to be bound to a nation-state, and they may indeed be capable of eliminating feudal revenants of 'the sovereign' in favor of fully modern constituent power and economic liberty. In particular, blockchains provide a ledger which enables a system of accounting (and hence a store of value), a means and medium of exchange, and a graphable network that can be used for deferred payments.

Whereas virtually all monetary theorists search for a money that contains all four functions—albeit in their preferred respective ratios—blockchain makes possible an unbundling of those functions into separate tokens that can nevertheless be systematically entwined. I would argue that many of the failures or egregious excesses of existing financial systems stems from the confused and mottled character of money having to perform all four functions at once. Financial systems can perhaps be more finely attuned if money's functions can achieve distinct incarnations.

3. POW vs POS

The thermodynamic work required for proof of work is fungible (computers of a given class are interchangeable); because this fungible capacity is limited by hardware production, PoW networks are—or risk becoming in their essence—winner-takes-all arms races. This is evident in BTC being mineable initially on any computer at all to required specialized and expensive equipment that requires constant updating to new, better, rarer equipment. This leads the Italian economists Massimo Amato and Luca Fantacci to argue in Fistfuls of Bitcoin that Bitcoin-the-payment-network is revolutionary, but that $BTC-the-token is a kind of vestigial artifact. Holding $BTC does not in any way grant the holder ownership over the Bitcoin payments network, only owning and operating and updating specialized hardware does that.

Proof of Stake networks promise, in theory, to reconcile this: PoS payment networks can retain BTC's resiliency through decentralization while granting their holders ownership of the network—what Jae rightly calls "general" rather than "limited partnership."

But PoS networks as they currently exist are almost indistinguishable from oligopoly and are always at risk of capture by moneyed interests; if they are not at this risk of capture by money, it is because some centralized authority, such as the PoS network's "foundation" or developers, control the majority of tokens and, through special delegations, control the flow of inflation. The deep irony of this latter case is that existing PoS networks become intrinsically bureaucratic, with unelected and unimpeachable validators effectively being selected/'made' by foundations, developers, VCs or other insiders, to essentially gatekeep the network's development. I will give two examples to illuminate:

a) In the case of $ATOM, the Hub followed this latter route and now has an entrenched bureaucracy of validators who are both well-capitalized (relative to other newcomers) and well-connected, allowing them to control other IBC chains. This unfortunately reproduces the 'winner take all' issue of BTC/PoW, but comes with the added risk that other ecosystem chains are 'Potemkin chains' with mal-aligned governors.

b) the $JUNO experiment—in theory, but not in practice—countervailed against the Hub model by radically decentralizing the network from the start with a substantial community airdrop... yet without any guardrails on governance, the Juno experiment simply fell prey to the first risk of PoS networks: capture by moneyed interests, often highly speculative and without substantial interest in developing the underlying assets (beyond, perhaps, what is necessary to pass it off to someone else).

4. A new Hub

The Cosmos SDK still (imo) provides the only meaningful tools to push a 'network state' experiment into production. Yes, the SDK has significant problems: the gov module is woefully inadequate (but Enterprise and DAO DAO have updated this in a powerful way within Cosmos), the validator system creates bureaucratic corruption (and requires a total rethink), etc. Moreover, the unary token system creates difficulties in all these facets.

What is needed is not simply the importation or recreation of fiat, but at least:

• a new Hub;
• a totally redeveloped validator system, perhaps severing validation from governance except on technical network issues.
• a binary, trinary, or even quaternary token system, with an uncapped inflationary token for security (staking rewards) and perhaps one-sided payments, a potentially recyclable Proof of Authority-style token for certain governance issues, and a hard capped store of value token, perhaps ultimately with a mutual credit graph network such as Cycles (but I believe a mutual credit system is a long way away from being important or viable);
• a legitimate (i.e., credibly neutral) and robust governance module, be that DAO DAO or something else; and
• a Hub-aligned system of one-sided payments.

As you can see, I'm excited about AtomOne because it has already recognized the importance of most of these points, especially Jae's comment about ATOM/PHOTON separation recognizing that PoW is in effect a dual-token network.

I'll write up my response to Jae's comments on the mark of the beast separately.