The Liquid Democracy Journal
on electronic participation, collective moderation, and voting systems
Issue 5
2017-05-11

The Origins of Liquid Democracy

by Jan Behrens, Berlin, May 11, 2017 other format: text version (UTF-8)

In late 2009, we (Jan Behrens, Axel Kistner, Andreas Nitsche, and Björn Swierczek) started to develop LiquidFeedback, a software for proposition development and decision making. [PLF] One of our goals was to provide a feasible implementation of Liquid Democracy. Back then, “Liquid Democracy” was only a vague concept to us that we stumbled upon in the orbit of the Pirate party movement in Germany. [Huwald] This article shall show that most of the ideas had already been thought of many decades ago, even though crucial details were devised around the millenium change.

Lewis Carroll's Principles of Parliamentary Representation from 1884

More than 130 years ago, Lewis Carroll (Charles Lutwidge Dodgson) described in his book “The Principles of Parliamentary Representation” the concept of giving candidates in an election a choice

It should be noted that the proceedings described in his book were specifically meant for elections of people (i.e. for representative democracy where members of a house of representatives are elected) and not for empowering the electorate to decide on certain issues themselves. Furthermore, the transfer of votes (according to his proposal) was carried out by candidates receiving those votes and not by voters.

His reasons to propose a transfer of votes by the candidates were to avoid a “waste of votes” in multi-winner elections where members of a house of representatives are being elected. Therefore, his proposal can be seen as alternative to “Single Transferrable Vote” (STV) systems, which had already been known at that time. Carroll justified his proposal by pointing out deficiencies found in the known methods to transfer votes using STV at that time, [Carroll, pp.30-32] and by claiming that preferential voting would be too complex for the “ignorant elector”:

»One great objection to this method is the confusion it would cause in the mind of an ignorant Elector, who, though quite able to name his favourite Candidate, would be utterly puzzled if told to arrange 5 or 6 names in order of merit.«
Lewis Carroll, 1884

Following the premise that neither excessive votes received by a candidate nor votes for those candidates who do not get elected should be lost, and arguing that it would not be feasible to let the voter decide who to transfer surplus and/or unused votes to, Carroll reasons that it must be “the candidate himself, for whom the votes have been given” who must have authority to dispose those votes. Guarding against skepticism, he explains that whoever trusts someone so far as to serve as a member of a house of representatives, he or she shouldn't have a problem to trust that person to transfer the vote in lieu thereof:

»The Elector must understand that, in giving his vote to [a candidate] ‘A’, he gives it him as his absolute property, to use for himself, or to transfer to other Candidates, or to leave unused. If he cannot trust the man, for whom he votes, so far as to believe that he will use the vote for the best, how comes it that he can trust him so far as to wish to return him as Member?«
Lewis Carroll, 1884

Obviously this argument could also serve in favor of transitive delegated voting, which was about to be invented more than 100 years later. [Note: There are more reasons to facilitate transitivity of delegations, as explained in [PLF, chapter 2].] However, Carroll's proposal didn't explicitly mention any transitive element, i.e. more than one step of transferring votes further. [Carroll, pp.36-40, pp.44-45]

While Lewis Carroll was probably the first person to propose empowering candidates to transfer their received votes, his proposal didn't contain any element to empower people to participate on particular issues themselves (which is arguably one of the most important aspects of Liquid Democracy, as described in the following).

Tullock's proposal in 1967

About 83 years later, Gordon Tullock proposed a hybrid of direct and representative democracy in form of a proportional representation scheme where each voter may decide to either represent him- or herself or to choose a representative. Each representative would be given a voting weight equal to the number of his or her voters but a voter could also elect him- or herself as their own representative (in which case they would receive a voting weight of at least 1). Tullock suggested to use the possibility of watching proceedings through broadcast and voting “by wire” (see also [Armytage14]) to allow an unlimited number of voters to represent themselves if they wish so. Tullock noted that only the existence of “computers” and “modern electronics” could make this possible:

»Indeed, until the development of the computer, the system would have been impracticable, which is probably the reason it has not heretofore been proposed. […] With modern electronics there is no necessity for all representatives to meet in the same hall, consequently there is no maximum on the number of representatives. Voting could easily be done by wire, and the proceedings could be broadcast. In the extreme case, there seems no reason why people who wish should not vote for themselves and then fill their days by casting their single vote for and against the various proposals.«
Gordon Tullock, 1967

Tullock's proposal might have been the first step towards a democracy where each voter may decide him- or herself whether to engage in decisions directly or through representation. It is thus a hybrid of direct and representative democracy. However, his proposal still differs from Liquid Democracy in at least two important ways: neither delegation by topic (i.e. nominating different people for different subject areas or issues) [PLF, subsection 2.2.2] nor the possibility to instantly revoke transferred voting weight [PLF, subsection 2.2.4] is part of Tullock's proposal. Tullock, however, did not expect his ideas to be the “best possible suggestions” but rather saw them as a potential step for “a better and more scientific political structure” in the future:

»New ideas always seem radical and bizarre. I would not claim that the new ideas I have advanced in these essays are the best possible suggestions. I hope, however, that they will play at least some role in the search for a better and more scientific political structure.«
Gordon Tullock, 1967

Two years later: Miller's Program for Direct and Proxy Voting

In 1969, James C. Miller foresaw in his publication “A program for direct and proxy voting in the legislative process” that within the next 20 or 30 years, every household would have a “console tied into a computer”. According to his ideas, such a computer console could not only be used for children to do their homework, making out grocery lists, or paying bills, but also for making political decisions such that each voter could decide on every issue:

»Some, in fact, have predicted that within 20 or 30 years every home will have a console tied into a computer upon which the children do their homework, the housewife will make out her grocery list, and the husband will pay the family's bills. Such a computer console also could be used to record political decisions, giving each voter an opportunity to cast his ballot on every issue and have it recorded through the machine.«
James C. Miller, 1969

Just like Tullock, Miller proposed that voters may use computer technology to vote on every single issue themselves or to delegate their vote to a representative if they wish to. But Miller further suggested that voters could determine themselves how long such a delegation shall be in effect:

»[…] instead of electing representatives peridically for a tenure of two years or more, why not allow citizens to vote directly or delegate proxy to someone else for as long as they like (which is, of course, analogous to stockholder voting schemes in large corporations).«
James C. Miller, 1969

While not foreseeing the difficulty regarding verifiability of such an electronic system, Miller did have some ideas about potential security measures, which were quite futuristic for that time:

»Safeguards, of course, would have to be installed so that no one could record decisions on the machine except its owner. For instance, a special metal key, a coded combination, or even a thumbprint might be required to operate the machine.«
James C. Miller, 1969

As of today, we know that measures such as “special metal keys”, coded combinations, or even retina scans will not be able to solve the problem of verfiable secret elections using computers (see also section 3.4 on the “Wahlcomputerproblem” in [PLF]). Nonetheless, (and disregarding the antiquated gender roles) Miller's proposal in that time is to be considered “visionary” (see also [Armytage14]).

As already said, the aspect of verifiability of electronic ballots is ignored by Miller at that time. Furthermore, his proposal (as well as Tullock's proposal) still differs in another way from what we call “Liquid Democracy” today: delegates who receive votes as proxies are not explicitly enabled to delegate their vote further. Such an extension isn't conflicting with Miller's proposal (and might even be considered part of the “stockholder voting schemes” he referenced) but to our knowledge hasn't been explicitly mentioned by him.

Miller, however, already proposed the ability to revoke a previously given delegation at any time. Ironically, the ability to instantly disempower (or empower) a representative was seen by Miller as a positive influence on the representative's freedom to vote his conscience because the dynamic representation scheme would allow representatives to “reverse their stand”:

»Under the proposal, the representative would be subject to instant recall by each and every voter. If a representative did not maintain the approval of those whose proxy votes he held, he would have them withdrawn and would find himself no longer a representative (unless, of course, he picked up proxy votes somewhere else). Such recall would be on a day-to-day or even an hour-to-hour basis. In a way, such a scheme probably would allow greater freedom for a representative to practice statesmanship and vote his conscience. Under the present system a representative must conform his general actions to the wishes of his (regional) supporters in order to be elected. But under the proposal, if a representative's ideas on policy issues changed and he conscientiously decided to reverse his stand, he could remain a representative by gathering proxy support from others holding the same general position.«
James C. Miller, 1969

Miller also considered the idea that voters might make their decision whether to vote directly or via proxy on a per-issue basis:

»Most voters […] would utilize some combination […], voting on major issues personally and delegating proxy to someone else for the minor decisions. Thus, the third feature of the proposal is a provision for proxy as well as direct voting.«
James C. Miller, 1969

The internet of the 90's: Rob Lanphier's Public Ballot Stewardship

The ideas from the 1960's reappeared on the internet around 1995. [Note: We didn't find a proof for the exact date of publication. The article published at the URL http://robla.net/1996/steward/ (accessed 2016-04-19) written by Rob Lanphier has a copyright notice dated 1995 [Lanphier] and has been cached on archive.org since 2005. James-Green Armytage states 1995 as the year of Lanphier's publication. [Armytage14]] Rob Lanphier proposed the “Public Ballot Stewardship”: a model for electronic democracy. [Lanphier] [Armytage14] He distinguished two forms of elections: general elections (using a secret ballot) and public elections. Depending on the impact of what is voted on, either a “general election” would be held, or – for more “mundane” things – a public ballot would be held, where everyone knows which way everyone else voted:

»General elections are what we now know as general elections. Private ballot, one person/one vote, you snooze you lose. Nothing fundemenatally different here from what we now call voting. These would be used on special issues like constitutional amendments, presidential elections, and other “big-ticket” items. Public elections are public ballot votes. Everyone knows which way everyone else voted, by name. They would be a matter of public record. These elections would be for the more “mundane” things, like budgets, minor bills, declaration of “National Boy Scouts Week”, etc.«
Rob Lanphier, 1995

In his proposal, the public elections provide a dynamic delegation system: Lanphier proposed that for each issue that is being decided in a public ballot, one may either decide to represent him- or herself or to choose a “steward” to vote on one's behalf. He furthermore notes two important aspects: one should be able to change the steward at any time, and for particular decisions one should be able to override the steward's vote by temporarily ceding the vote from the steward and voting directly. [Lanphier] Lanphier even proposed to allow different delegations in different subject areas:

»Maybe, instead of one “body”, there could be several congresses, each with assigned powers of their own. One congress dedicated to the interior, one to defense, one to education. People could pick individual stewards for each, or choose one to handle all. Perhaps this would be done on an individual basis, where the steward has trusted advisors that actually choose the vote in their given specialty.«
Rob Lanphier, 1995

Lanphier's proposal doesn't explicitly state that delegates may further delegate their votes (see [PLF, chapter 2] for transitive delegation). The cited “trusted advisors” could, however, be seen as a precursor to what would later be proposed in the 2000's by Bryan Ford and others.

There was one other important aspect which Lanphier addressed in his publication: the internet (as of 1995) didn't provide means to identify its users. Lanphier pointed out that it would be necessary to verify that one person gets exactly one vote (and not two, three, or more, by creating multiple accounts). He assumed, however, that this problem would be solved in the near future:

»The main problem facing electronic voting on the Internet is verifying that one person gets one vote, and that all people are represented (even those without Internet access). Verification of identity is a problem that plagues many applications on the Internet (such as making purchases on the net, or filing taxes on the net), and so this one will likely be solved regardless of whether electronic voting makes it an issue.«
Rob Lanphier, 1995

Similar considerations can also be found in our book “The Principles of LiquidFeedback”, [PLF, subsection 6.1.1] even though we were not aware of his publication at the time of writing our book. It should be noted that Lanphier's optimistic views in regard to emerging means of identity verification on the internet would fail to come true, at least for another 20 years.

Bryan Ford's Delegative Democracy

In the early 2000's [Note: We didn't find a proof for the exact date of publication. The PDF published at http://www.brynosaurus.com/deleg/deleg.pdf carries a date of May 15, 2002 and has been cached on archive.org since 2005. A link to the document as well as Ford's pages on Delegative Voting have been cached on archive.org since 2004. See also [Armytage14], where Ford's ideas are dated 2002.], Bryan Ford proposed two ideas which he calls “Delegative Voting” [FordDV] and “Delegative Democracy” [FordDD] of which the former can be seen as further development of Lewis Carroll's ideas (a method to avoid wasted votes when electing individuals without preferential ballots) and the latter contains elements akin to Lanphier's proposal. However, instead of following Lanphier's idea to facilitate two different forms of ballots (secret and public) for different kinds of decisions, Ford demanded “privacy of the individuals” and a transitive delegation system at the same time for the same decisions. He named the following 6 basic principles to describe his form of “Delegative Democracy”:

As for the last point, Ford added the important property of transitivity to the delegation model which wasn't mentioned in the previous proposals of Tullock, Miller, and Lanphier. However, in regard to the Wahlcomputerproblem (see [PLF, chapter 3]), he makes a big step backward (when compared to Lanphier) because his proposal may not be applied to electronic systems without losing another imporant property of democractic processes, namely verifiability.

Following Ford's step backward, the idea of privacy of the individual and accountability of the delegate would later be described by Hardt and Lopes [Hardt&Lopes] as a possible solution regarding their “Golden Rule of Liquid Democracy” (which, we think, is a dangerous euphemism). [Editorial4]

Even though we do not consider private electronic ballots to be part of the concept of Liquid Democracy or to meet democratic standards at all [Editorial4], Bryan Ford may have been the first person to add transitive delegations as the last missing piece to describe what is called “Liquid Democracy” today. [Note: Later, other persons claimed to be the “inventors” of Liquid Democracy. However, we can not confirm their claims. [Deseriis] See also [QA010]. We would also like to note that most ideas were already formulated in 1967, 1969, and 1995 by Tullock, Miller, and Lanphier respectively.]

Ford furthermore described three possible “extensions” to the delegation model: “backup choices”, “split delegation”, and “restricted and transitive delegation”. Bryan Ford didn't provide any notes on implementation in his paper, neither for a system fulfilling those 6 basic principles listed above nor for these three extensions. Section 4 of his paper just reads, “Implementing Delegative Democracy : Under Construction”. [FordDD] As it could be shown later in [PLF, subsection 2.4.2] and [PD], two of his extensions would add certain unwanted properties to his system (negative voting weight and/or unequal treatment of the participants).

2004: Combining transitive delegations with preferential voting on issues

James Green-Armytage described in 2004 [Note: An early draft of his ideas is available on archive.org. [Armytage04]] and 2005 another system that is incorporating transitive delegations for decision making. [Armytage05] He called his system “Direct Democracy by Delegable Proxy”, which consists of the following two “fundamental elements”:

Opposed to the visionary views in 1967 and 1969, Armytage didn't see the internet as an application field yet:

»Should the votes [should] take place over the internet, or only at controlled polling stations? The internet poses problems of security and problems of equal access, so I suggest that official polling stations are a preferable venue. The voter interface should be electronic (paper ballots would probably just be too clumsy for this system), and every effort should be made to assure that the votes are being counted accurately.«
James Green-Armytage, 2005

It should be noted that electronic voting machines do not solve the problems of verifiable electronic voting either. [PLF, section 3.4] [BVerfG] [CCC] However, combining transitive delegations [PLF, chapter 2] with a preferential voting system on proposals [PLF, section 4.12] are some of the key elements found in our software LiquidFeedback, even though we were not aware of Armytage's website at the time of creation of our software. Armytage also proposed that delegations shall be “issue-specific” in order to allow for a delegation to experts in each field, and he explained the importance of transitivity:

»One reason this might be good is that it would allow voters to indicate as proxies people who are knowledgeable in the field that a specific issue relates to. For example, if the issue is relevant to ecology, then a voter might indicate an ecologist as their proxy for that issue, or a staff member at an NGO that deals with the environment. Or, rather than being a matter of a field of study, a voter may delegate his vote to someone whom he knows has educated themselves well about that issue in particular. For example, if the issue is choosing between different versions of a trade bill and the voter knows someone who has read all of the different versions personally. Even if most voters would not know such a person, their proxies and their proxies' proxies might.«
James Green-Armytage, 2004

The publication of LiquidFeedback in 2009

In the year 2009, Jan Behrens, Axel Kistner, Andreas Nitsche, and Björn Swierczek presented “LiquidFeedback”, which doesn't only incorporate the ideas regarding Liquid Democracy as described above [Note: with the exception of secrecy as explained in chapter 3 of [PLF]], but also includes other features such as proportional minority representation and preferential voting which go beyond the ideas of Liquid Democracy. A more detailed description of LiquidFeedback can be found in the book, “The Principles of LiquidFeedback”. [PLF]

Summary

The origins of Liquid Democracy date back as far as 1884, [Carroll] though most core elements were not foreseen until 1969, when James C. Miller published “A program for direct and proxy voting in the legislative process”. [Miller] In the year 1995, the idea of vote delegation was re-thought in the context of the emerging use of the internet. [Lanphier] The element of transitivity was brought up by several people after the millenium change.

The term “Liquid Democracy” became popular in Germany when the Pirate party movement proposed to overcome the limitations of direct and representative democracy. [Huwald] Even though the Pirate party arguably never succeeded in establishing a sustainable application of Liquid Democracy, they helped the idea to gain attention by public media.

LiquidFeedback was created independently of any particular party as a feasible software solution to help political parties or other organizations to implement Liquid Democracy within their organization. Beside the concepts of Liquid Democracy, a particular set of rules of procedure was combined with the ideas of Liquid Democracy to allow users to engage in a collective decision-making process where all participants are treated equally. [PLF] With the publication of LiquidFeedback, it is possible to apply Liquid Democracy to real-world scenarios.

[Armytage04] James Green-Armytage: “A Proposal for Direct Democracy Based on a Non-Binding Proxy System”, archived June 3, 2004, available at http://web.archive.org/web/20040603225236/http://fc.antioch.edu/~jarmyta@antioch-college.edu/voting_methods/proxy.htm (accessed 2016-04-19) (referenced at: a b)
[Armytage05] James Green-Armytage: “Direct Democracy by Delegable Proxy”, 2005. Accessed at http://inside.bard.edu/~armytage/proxy2005.pdf (on 2016-04-19, dead link as of 2017-05-01), also archived as HTML on April 28, 2005, available at http://web.archive.org/web/20050428192452/http://fc.antioch.edu/~james_green-armytage/vm/proxy.htm (accessed 2016-04-19) (referenced at: a b)
[Armytage14] James Green-Armytage: “Direct Voting and Proxy Voting”, October 26, 2014. Accessed at http://inside.bard.edu/~armytage/proxy.pdf (on 2016-04-19, dead link as of 2017-05-01), archived December 10, 2014, available at http://web.archive.org/web/20141210071045/http://inside.bard.edu:80/~armytage/proxy.pdf (accessed 2017-05-10) (referenced at: a b c d e)
[BVerfG] Decision of German Federal Constitutional Court: “BVerfG, 2 BvC 3/07”, March 3, 2009, Absatz Nr. 1-163. Published by Bundesverfassungsgericht, Schlossbezirk 3, 76131 Karlsruhe, Germany. Available at http://www.bverfg.de/entscheidungen/cs20090303_2bvc000307.html (referenced at: a)
[Carroll] Lewis Carroll (Charles Lutwidge Dodgson): “The Principles of Parliamentary Representation”, 1884. Available at http://babel.hathitrust.org/cgi/pt?id=uva.x000676622 or https://archive.org/details/ThePrinciplesOfParliamentaryRepresentation (referenced at: a b c d e f)
[CCC] Constanze Kurz, Frank Rieger, Rop Gonggrijp: “Beschreibung und Auswertung der Untersuchungen an NEDAP-Wahlcomputern”, May 30, 2007. Published by Chaos Computer Club e. V., Humboldtstraße 53, 22083 Hamburg, Germany. Available at http://wahlcomputer.ccc.de/doku/nedapReport54.pdf (accessed 2016-04-19) (referenced at: a)
[Deseriis] Jan Behrens, Axel Kistner, Andreas Nitsche, Björn Swierczek, Marco Deseriis: “Liquid democracy, its challenges and its forebears”, October 9, 2015. Published at https://www.opendemocracy.net/can-europe-make-it/jan-behrens-axel-kistner-andreas-nitsche-bj%C3%B6rn-swierczek-marco-deseriis/liquid-de (accessed at 2016-04-19) (referenced at: a)
[Editorial4] Behrens, Kistner, Nitsche, Swierczek: Editorial of “The Liquid Democracy Journal on electronic participation, collective moderation, and voting systems”, Issue 4, July 28, 2015, pp. 3-5. ISSN 2198-9532. Published by Interaktive Demokratie e. V., available at http://www.liquid-democracy-journal.org/issue/4/The_Liquid_Democracy_Journal-Issue004-00-Editorial.html See also corrigendum in Issue 5, May 11, 2017, p. 4. (referenced at: a b)
[FordDD] Bryan Ford: “Delegative Democracy”, May 15, 2002. Available at http://www.brynosaurus.com/deleg/deleg.pdf and also archived March 6, 2005, available at http://web.archive.org/web/20050306143310/http://www.brynosaurus.com/deleg/deleg.pdf (referenced at: a b)
[FordDV] Bryan Ford: “Delegative Voting”, October 21, 2002. Available at http://www.brynosaurus.com/log/2002/1021-DelegativeVoting/ (accessed 2016-04-19) and also archived August 17, 2004, available at http://web.archive.org/web/20040817195022/http://www.brynosaurus.com/deleg/elect/ (accessed 2016-04-19) (referenced at: a)
[Hardt&Lopes] Steve Hardt, Lia C. R. Lopes: “Google Votes: A Liquid Democracy Experiment on a Corporate Social Network”, June 5, 2015. In “Defensive Publications Series”, 2015, on Technical Disclosure Commons, available at http://www.tdcommons.org/dpubs_series/79 (accessed 2017-05-10)
[Huwald] Jan Huwald interviewed by Peter Mühlbauer: “Warum Partei und nicht Religion?”, September 12, 2007. Published by Heise Medien GmbH & Co. KG, Telepolis, available at http://www.heise.de/tp/artikel/26/26173/1.html (accessed 2016-04-19) (referenced at: a b)
[Lanphier] Rob Lanphier: “A Model for Electronic Democracy?” Published at http://robla.net/1996/steward/ (accessed 2016-04-19) and also archived on October 27, 2005, available at http://web.archive.org/web/20051027051842/http://robla.net/1996/steward/ (accessed 2016-04-19) (referenced at: a b c d e f g)
[Miller] James C. Miller III: A program for direct and proxy voting in the legislative process. In “Public Choice”, Vol. 7, Issue 1, pp. 107-113. ISSN 0048-5829 (Print), 1573-7101 (Online). Published by Springer, available at http://link.springer.com/article/10.1007/BF01718736 (referenced at: a b c d e f)
[PD] Jan Behrens and Björn Swierczek: Preferential Delegation and the Problem of Negative Voting Weight. In “The Liquid Democracy Journal on electronic participation, collective moderation, and voting systems”, Issue 3, January 23, 2015, pp. 6-34. ISSN 2198-9532. Published by Interaktive Demokratie e. V., available at http://www.liquid-democracy-journal.org/issue/3/The_Liquid_Democracy_Journal-Issue003-01-Preferential_Delegation_and_the_Problem_of_Negative_Voting_Weight.html (referenced at: a)
[PLF] Behrens, Kistner, Nitsche, Swierczek: “The Principles of LiquidFeedback”. ISBN 978-3-00-044795-2. Published January 2014 by Interaktive Demokratie e. V., available at http://principles.liquidfeedback.org/ (referenced at: a b c d e f g h i j k l m n o)
[QA010] The developers of LiquidFeedback (Behrens, Kistner, Nitsche, Swierczek): Readers of the Journal Asked – LiquidFeedback Developers Answer (#2), Question & Answer 10. In “The Liquid Democracy Journal on electronic participation, collective moderation, and voting systems”, Issue 5, May 11, 2017, pp. 36-37. ISSN 2198-9532. Published by Interaktive Demokratie e. V. (referenced at: a)
[Tullock] Gordon Tullock: Proportional representation. In “Toward a Mathematics of Politics”, pp. 144-157. Library of Congress Catalog No. 67-27335. Published by The University of Michigan Press (U.S.A.) and Ambassador Books Limited (Rexdale, Canada). (referenced at: a b)