Crypto Anarchy and Virtual Communities

Crypto Anarchy and Virtual Communities

Timothy C. May

Modern Cryptography

The past two decades have produced a revolution in cryptography
(crypto, for short)—the science of the making of ciphers and codes.
Beyond just simple ciphers, useful mainly for keeping communications
secret, modern crypto includes diverse tools for authentication of mes-
sages, for digital time stamping of doc**ents, for hiding messages in
other doc**ents (steganography), and even for schemes for digital
cash.

Public key cryptography, the creation of Diffie and Hellman, has
dramatically altered the role of crypto. Coming at the same time as the
wh***sale conversion to computer networks and worldwide communi-
cations, it has been a key element of security, confidence, and success.
The role of crypto will only become more important over the coming
decades. Pretty Good Privacy (PGP) is a popular version of the algorithm
developed by Rivest, Shamir, and Adleman (known, of course, as RSA).
The RSA algorithm was given a patent in the United States, though not
in any European countries, and is licensed commercially.1

These tools are described in detail in various texts and conference
proceedings and are not the subject of this chapter.2 The focus here is on
the implications of strong crypto for cyberspace, especially on virtual
communities. Mention should be made of the role of David Chaum in
defining the key concepts here. In several seminal papers,3 Chaum intro-
duced the ideas of using public key cryptography methods for anony-
mous, untraceable electronic mail, for digital money systems in which
spender identity is not revealed, and in schemes related to these. (I make
no claims that Chaum agrees with my conclusions about the political
and socioeconomic implications of these results.)


Virtual Communities

Notes: cyberspace, Habitat, VR, Vinge, etc. Crypto holds up the “walls”
of these cyberspatial realities. Access control, access rights, modification
privileges.

Virtual communities are the networks of individuals or groups that
are not necessarily closely connected geographically. The “virtual” is
meant to imply a nonphysical linking but should not be taken to mean
that these are any less community like than are conventional physical
communities.

Examples include churches, service organizations, clubs, criminal
gangs, cartels, fan groups, etc. The Catholic Church and the Boy Scouts
are both examples of virtual communities that span the globe, transcend
national borders, and create a sense of allegiance, of belonging, and a
sense of community. Likewise, the Mafia is a virtual community (with its
enforcement mechanisms, its own extralegal rules, etc.) Lots of other
examples: Masons, Triads, the Red Cross, Interpol, Islam, Judaism,
Mormons, Sindero Luminoso, the IRA, drug cartels, terrorist groups,
Aryan Nation, Greenpeace, the Animal Liberation Front, and so on.

There are undoubtedly many more such virtual communities than there
are nation-states, and the ties that bind them are for the most part much
stronger than are chauvinist nationalist emotions. Any group in which
the common interests of the group, be it a shared ideology or a particu-
lar interest is enough to create a cohesive community.

Corporations are another prime example of a virtual community, hav-
ing scattered sites, private communication channels (generally inacces-
sible to the outside world, including the authorities), and their own goals
and methods. In fact, many “cyberpunk” (not cypherpunk) fiction
authors make a mistake, I think, in assuming the future world will be
dominated by transnational megacorporate “states.” In fact, corpora-
tions are just one example of many of such virtual communities that will
be effectively on a par with nation states. (Note especially that any laws
designed to limit use of crypto cause immediate and profound problems
for corporations and that countries like France and the Philippines,
which have attempted to limit the use of crypto, have mostly been
ignored by corporations. Any attempts to outlaw crypto will produce a
surge of sudden “incorporations,” thus gaining for the new corporate
members the aegis of corporate privacy.) In an academic setting, “invis-
ible colleges” are the communities of researchers.

These virtual communities typically are “opaque” to outsiders. At-
tempts to gain access to the internals of these communities are rarely suc-
cessful. Law-enforcement and intelligence agencies (such as the National
Security Agency in the United States, Chobetsu in Japan, SDECE in
France, and so on) may infiltrate such groups and use electronic surveil-
lance (ELINT) to monitor these virtual communities. Not surprisingly,
these communities have been early adopters of encryption technology,
ranging from scrambled cellphones to full-blown PGP encryption.4

The use of encryption by “evil” groups—such as child pornographers,
terrorists, abortionists, and abortion protesters—is cited by those who
wish to limit civilian access to crypto tools. We call these groups the
“Four Horsemen of the Infocalypse,” as they are so often cited as the
reason that ordinary citizen units of the nation state should not have
access to crypto.

This is clearly a dangerous argument to make, for various good rea-
sons. The basic right of free speech is the right to speak in a language
one’s neighbors or governing leaders may not find comprehensible—
encrypted speech. There’s not enough space here to go into the many
good arguments against a limit on access to privacy, communications
tools, and crypto.

The advent of full-featured communications systems for computer-
mediated virtual communities will have even more profound impli-
cations. MUDs and MOOs (multi-user domains, etc.) and 3D virtual
realities are one avenue, and text-centric Net communications are
another. (Someday, soon, they’ll merge, as described in Vernor Vinge’s
prophetic 1980 novella, True Names.)


Observability and Surveillance

An interesting way to view issues of network visibility is in terms of the
transparency of nodes and links between nodes. Transparent means vis-
ible to outsiders, perhaps those in law enforcement or the intelligence
community. Opaque means not transparent, not visible. A postcard is
transparent; a sealed letter is opaque. PGP inventor Phil Zimmermann
has likened the requirement for transparency to being ordered to use
postcards for all correspondence, with encryption the equivalent of an
opaque envelope (envelopes can be opened, of course, and long have
been).

Transparent links and nodes are the norm in a police state, such as
the former Soviet Union, Iraq, China, and so forth. Communications
channels are tapped, and private use of computers is restricted. (This is
becoming increasingly hard to do, even for police states; many cite the
spread of communications options as a proximate cause of the collapse
of communism in recent years.)

There are interesting “chemistries” or “algebras” of transparent ver-
sus opaque links and nodes. What happens if links must be transparent
but nodes are allowed to be opaque? (The answer: the result is the same
as if opaque links and nodes were allowed—that is, the full implications
of strong crypto. Hence, any attempt to ban communications crypto
while still allowing private CPUs to exist. . . .)

If Alice and Bob are free to communicate, and to choose routing paths,
then Alice can use “crypto arbitrage” (a variation on the term, “reg-
ulatory arbitrage,” the term Eric Hughes uses to capture this idea of
moving transactions to other jurisdictions) to communicate with sites—
perhaps in other countries—that will perform as she wishes. This can
mean remailing, mixing, etc. As an example, Canadian citizens who are
told they cannot access information on the H***lka-Teale murder case
(a controversial case in which the judge has ordered the media in Canada
and entering Canada not to discuss the gory details) nevertheless have
a vast array of options, including using telnet, gopher, ftp, the Web,
etc., to access sites in many other countries or even in no country in
particular.

Most of the consequences described here arise from this chemistry of
links and nodes: unless nearly all node and links are forced to be trans-
parent, including links to other nations and the nodes in those nations,
then private communication can still occur. Crypto anarchy results.


Crypto Anarchy

“The Net is an anarchy.” This truism is the core of crypto anarchy—no
central control, no ruler, no leader (except by example or reputation), no
“laws.” No single nation controls the Net, no administrative body sets
policy. The Ayatollah in Iran is as powerless to stop a newsgroup—
alt.wanted.moslem.women or alt.wanted.moslem.gay come to mind—he
doesn’t like as the president of France is as powerless to stop, say, the
abuse of the French language in soc.culture.french. Likewise, the CIA
can’t stop newsgroups, sites, or Web pages that give away its secrets. At
least not in terms of the Net itself. What non-Net steps might be taken
are left as an exercise for the paranoid and the cautious.

This essential anarchy is much more common than many think.
Anarchy (the absence of a ruler telling another person what to do) is
common in many walks of life—choosing books to read, movies to see,
friends to socialize with, and so on. Anarchy does not mean complete
freedom (we can, after all, read only the books that someone has written
and had published), but it does mean freedom from external coercion.
Anarchy as a concept, though, has been tainted by other associations.

First, the anarchy here is not the anarchy of popular conception—
lawlessness, disorder, and chaos. Nor is it the bomb-throwing anarchy
of the nineteenth century “black” anarchists, usually associated with
Russia and labor movements. Nor is it the “black flag” anarchy of anar-
cho-syndicalism and writers such as Proudhon. Rather, the anarchy
being spoken of here is the anarchy of “absence of government” (liter-
ally, “an arch,” without a chief or head).

This is the same sense of anarchy used in anarchocapitalism, the liber-
tarian free-market ideology that promotes voluntary, uncoerced eco-
nomic transactions.5 I devised the term crypto anarchy as a pun on
crypto, meaning “hidden,” on the use of “crypto” in combination with
political views (as in Gore Vidal’s famous charge to William F. Buckley:
“You’re crypto fascist!”) and of course because the technology of crypto
makes this form of anarchy possible. The first presentation of this term
was in a 1988 “Manifesto,” whimsically patterned after another famous
manifesto.6 Perhaps a more popularly understandable term, such as
“cyber liberty,” might have some advantages, but crypto anarchy has its
own charm, I think.

And anarchy in this sense does not mean that local hierarchies don’t
exist or that no rulers exist. Groups outside the direct control of local
governmental authorities may still have leaders, rulers, club presidents,
elected bodies, etc. Many will not, though.
Politically, virtual communities outside the scope of local governmen-
tal control may present problems of law enforcement and tax collection.
(Some of us like this aspect.) Avoidance of coerced transactions can
mean avoidance of taxes, avoidance of laws saying who one can sell to
and who one can’t, and so forth. It is likely that many will be unhappy
that some are using cryptography to avoid laws designed to control
behavior.

National borders are becoming more transparent than ever to data. A
flood of bits crosses the borders of most developed countries’ phone
lines, cables, fibers, satellite up/downlinks, and millions of diskettes,
tapes, CDs, etc. Stopping data at the borders is less than hopeless.
Finally, the ability to move data around the world at will, the ability
to communicate to remote sites at will, means that a kind of “regulatory
arbitrage” can be used to avoid legal roadblocks. For example, when
remailing into the United States from a site in the Netherlands, whose
laws apply? (If one thinks that U.S. laws should apply to sites in the
Netherlands, then does Iraqi law apply in the United States? And so on.)

This regulatory arbitrage is also useful for avoiding the welter of laws
and regulations that operations in one country may face, including the
“deep pockets” lawsuits so many in the United States face. Moving oper-
ations on the Net outside a litigious jurisdiction is one step to reduce this
business liability. Like Swiss banks, but different.
True Names and Anonymous Systems
Something needs to be said about the role of anonymity and digital pseud-
onyms. This is a topic for an essay unto itself, of course.
Are true names really needed? Why are they asked for? Does the na-
tion state have any valid reason to demand they be used?

People want to know who they are dealing with, for psychological/
evolutionary reasons and to better ensure traceability should they need
to locate a person to enforce the terms of a transaction. The purely
anonymous person is perhaps justifiably viewed with suspicion.
And yet pseudonyms are successful in many cases. We rarely know
whether someone who presents himself by some name is “actually” that
person. Authors, artists, performers, etc., often use pseudonyms. What
matters is persistence and nonforgeability. Crypto provides this.

On the Cypherpunks7 mailing list, well-respected digital pseudonyms
have appeared and are thought of no less highly than their “real” col-
leagues are.

The wh*** area of digitally authenticated reputations, and the “repu-
tation capital” that acc**ulates or is affected by the opinions of others,
is an area that combines economics, game theory, psychology, and
expectations. A lot more study is needed. It is unclear if governments
will move to a system of demanding “Information Highway Driver’s
Licenses,” figuratively speaking, or how systems like this could ever be
enforced. (The chemistry of opaque nodes and links, again.)


Examples and Uses

It surprises many people that some of these uses are already being inten-
sively explored. Anonymous remailers are used by tens of thousands of
persons—and perhaps abused.8 And of course encryption, via RSA, PGP,
etc., is very common in some communities (hackers, Net users, freedom
fighters, white separatists, etc. . . . I make no moral judgments here about
people who use these methods).
Remailers are a good example to look at in more detail. There are two
current main flavors of remailers:
. Cypherpunk-style remailers process text messages to redirect mail to
other sites, using a command syntax that allows arbitrary nesting of
remailing (as many sites as one wishes) with PGP encryption at each level
of nesting.
. Julf-style remailers are based on the original work of Karl Kleinpaste
and are operated/maintained by Julf Helsingius in Finland. No encryp-
tion, and only one such site at present. (This system has been used ex-
tensively for messages posted to the Usenet and is basically successful.

The model is based on operator trustworthiness and his location in
Finland, beyond the reach of court orders and subpoenas from most
countries.)

The Cypherpunks remailers currently number about twenty, with
more being added every month. There is no reason not to expect hun-
dreds of such remailers in a few years. One experimental “information
market” is BlackNet, a system that appeared in 1993 and that allows
fully anonymous, two-way exchanges of information of all sorts. There
are reports that U.S. authorities have investigated BlackNet because of its
presence on networks at Defense Department research labs. Not much
they can do about it, of course, and more such entities are expected.
The implications for espionage are profound and largely unstoppable.
Anyone with a home computer and access to the Net or Web, in various
forms, can use these methods to communicate securely, anonymously, or
pseudonymously and with little fear of detection. “Digital dead drops”
can be used to post information obtained, far more securely than the old
physical dead drops (no more messages left in Coke cans at the bases of
trees on remote roads).

Whistleblowing is another growing use of anonymous remailers, with
folks fearing retaliation using remailers to publicly post information.
(Of course, there’s a fine line between whistle blowing, revenge, and
espionage.)

Data havens for the storage and marketing of controversial informa-
tion is another area of likely future growth. Nearly any kind of infor-
mation—medical, religious, chemical, etc., is illegal or proscribed in one
or more countries, so those seeking this illegal information will turn to
anonymous messaging systems to access and perhaps purchase this infor-
mation with anonymous digital cash. This might include credit data-
bases, deadbeat renter files, organ bank markets, etc. (These are all things
which have various restrictions on them in the United States. For exam-
ple, one cannot compile credit databases or lists of deadbeat renters
without meeting various restrictions—a good reason to move them into
cyberspace or at least outside the United States and then sell access
through remailers.) Matching buyers and sellers of organs is another
such market with a huge demand (life and death) but various laws tightly
controlling such markets.


Digital cash efforts. A lot has been written about digital cash.9 David
Chaum’s company, DigiCash, has the most interesting technology and
has recently begun market testing. Stefan Brands may or may not have a
competing system that gets around some of Chaum’s patents. (The atti-
tude crypto anarchists might take about patents is another topic for dis-
cussion. Suffice it to say that patents and other intellectual property
issues continue to have relevance in the practical world, despite erosion
by technological trends.) Credit card–based systems, such as the First
Virtual system, are not exactly digital cash, in the Chaumian sense of
blinded notes, but they offer some advantages the market may find use-
ful until more advanced systems are available. I expect to see many more
such experiments over the next several years, and some of them will
likely be market successes.


Commerce and Colonization of Cyberspace

How will these ideas affect the development of cyberspace? “You can’t
eat cyberspace” is a criticism often leveled at argument about the role
of cyberspace in everyday life. The argument made is that money and
resources “acc**ulated” in some future (or near future) cyberspatial
system will not be able to be “laundered” into the real world. Even such
a prescient thinker as Neal Stephenson, in Snow Crash,10 had his pro-
tagonist a vastly wealthy man in “the Multiverse” but a near pauper in
the physical world.

This is implausible for several reasons. First, we routinely see transfers
of wealth from the abstract world of stock tips, arcane consulting knowl-
edge, etc., to the real world. “Consulting” is the operative word. Second,
a variety of means of laundering money, via phony invoices, uncollected
loans, art objects, etc., are well-known to those who launder money. . . .
These methods, and more advanced ones to come, are likely to be used
by those who wish their cyberspace profits moved into the real world.
(Doing this anonymously and untraceably is another complication.
There may be methods of doing this. Proposals have looked pretty solid,
but more work is needed.)

The World Wide Web is growing at an explosive pace. Combined with
cryptographically protected communication and digital cash of some
form (and there are several being tried), this should produce the long-
awaited colonization of cyberspace. Most Net and Web users already
pay little attention to the putative laws of their local regions or nations,
apparently seeing themselves more as members of various virtual com-
munities than as members of locally governed entities.

This trend is accelerating.

Most important, information can be bought and sold (anonymously,
too) and then used in the real world. There is no reason to expect that
this won’t be a major reason to move into cyberspace.
Implications
I’ve touched on the implications in several places. Many thoughtful peo-
ple are worried about some of the possibilities made apparent by strong
crypto and anonymous communication systems. Some are proposing
restrictions on access to crypto tools. The recent debate in the United
States over Clipper and other key escrow systems shows the strength of
emotions generated by this issue.

Abhorrent markets may arise. For example, anonymous systems and
untraceable digital cash have some obvious implications for the arrang-
ing of contract killings and such. (The greatest risk in arranging such hits
is that physical meetings expose the buyers and sellers of such services to
stings. Crypto-anarchy lessens, or even eliminates, this risk, thus lower-
ing transaction costs. The risks to the actual triggermen are not lessened,
but this is a risk the buyers need not worry about. Think of anonymous
escrow services which hold the digital money until the deed is done. Lots
of issues here. It is unfortunate that this area is so little-discussed. . . .
People seem to have an aversion for exploring the logical consequences
in such areas.) The implications for corporate and national espionage
have already been touched upon. Combined with liquid markets in infor-
mation, this may make secrets much harder to keep. (Imagine a “Digital
Jane’s,” after the military weapons handbooks, anonymously compiled
and sold for digital money, beyond the reach of various governments
which don’t want their secrets told.)

New money-laundering approaches are another area to explore.
Something that is inevitable is the increased role of individuals, lead-
ing to a new kind of elitism. Those who are comfortable with the tools
described here can avoid the restrictions and taxes that others cannot. If
local laws can be bypassed technologically, the implications are pretty
clear.

The implications for personal liberty are of course profound. No
longer can nation-states tell their citizen-units what they can have access
to, not if these citizens can access the cyberspace world through anony-
mous systems.


How Likely?

I am making no bold predictions that these changes will sweep the
world anytime soon. Most people are ignorant of these methods, and
the methods themselves are still under development. A wh***sale con-
version to “living in cyberspace” is just not in the cards, at least not in
the next few decades. But to an increasingly large group, the Net is real-
ity. It is where friends are made, where business is negotiated, where
intellectual stimulation is found. And many of these people are using
crypto-anarchy tools. Anonymous remailers, message pools, informa-
tion markets. Consulting via pseudonyms has begun to appear and
should grow. (As usual, the lack of a robust digital cash system is slow-
ing things down.)

Can crypto-anarchy be stopped? Although the future evolution in
unclear, as the future almost always is, it seems unlikely that present
trends can be reversed:
. Dramatic increases in bandwidth and local, privately owned computer
power,
. Exponential increase in the number of Net users,
. Explosion in degrees of freedom in personal choices, tastes, wishes, and
goals, and
. Inability of central governments to control economies, cultural trends,
and so on.11

The Net is integrally tied to economic transactions, and no country can
afford to “disconnect” itself from it. (The U.S.S.R. couldn’t do it, and
they were light-years behind the U.S., European, and Asian countries.)
And in a few more years, no hope of limiting these tools at all, something
the U.S. F.B.I. has acknowledged.12

Technological Inevitability: These tools are already in widespread use,
and only draconian steps to limit access to computers and communi-
cations channels could significantly impact further use. (Scenarios for
restrictions on private use of crypto.)

As John Gilmore has noted, “The Net tends to interpret censorship as
damage, and routes around it.” This applies as well to attempts to leg-
islate behavior on the Net. (The utter impossibility of regulating the
worldwide Net—with entry points in more than a hundred nations, with
million of machines—is not yet fully recognized by most national gov-
ernments. They still speak in terms of “controlling” the Net, when in fact
the laws of one nation generally have little use in other countries.)

Digital money in its various forms is probably the weakest link at this
point. Most of the other pieces are operational, at least in basic forms,
but digital cash is (understandably) harder to deploy. Hobbyist or “toy”
experiments have been c**bersome, and the “toy” nature is painfully
obvious. It is not easy to use digital cash systems at this time (“To use
Magic Money, first create a client . . .”), especially as compared to the
easily understood alternatives.13 People are understandably reluctant to
entrust actual money to such systems. And it’s not yet clear what can be
bought with digital cash (a chicken or egg dilemma that is likely to be
resolved in the next several years). Digital cash, digital banks, etc., are a
likely target for legislative moves to limit the deployment of crypto anar-
chy and digital economies. Whether through banking regulation or tax
laws, it is not likely that digital money will be deployed easily (“Kids,
don’t try this at home!”).

Some of the current schemes may also incorporate methods for re-
porting transactions to the tax authorities and may include “software
key escrow” features that make transactions fully or partly visible to
authorities.


Conclusions

Strong crypto provides new levels of personal privacy, all the more
important in an era of increased surveillance, monitoring, and the temp-
tation to demand proofs of identity and permission slips. Some of the
“credentials without identity” work of Chaum and others may lessen
this move toward a surveillance society.

The implications are, as I see it, are that the power of nation states will
be lessened, tax collection policies will have to be changed, and eco-
nomic interactions will be based more on personal calculations of value
than on societal mandates.

Is this a Good Thing? Mostly yes. Crypto anarchy has some messy
aspects, of this there can be little doubt. From relatively unimportant
things like price fixing and insider trading to more serious things like
economic espionage, the undermining of corporate knowledge owner-
ship, to extremely dark things like anonymous markets for killings.
But let’s not forget that nation states have, under the guise of protect-
ing us from others, killed more than 100 million people in this century
alone. Mao, Stalin, Hitler, and Pol Pot, just to name the most extreme
examples. It is hard to imagine any level of digital contract killings ever
coming close to nationstate barbarism. (But I agree that this is something
we cannot accurately speak about; I don’t think we have much of a
choice in embracing crypto anarchy or not, so I choose to focus on the
bright side.) It is hard to argue that the risks of anonymous markets and
tax evasion justify worldwide suppression of communications and
encryption tools. People have always killed each other, and governments
have not stopped this (arguably, they make the problem much worse, as
the wars of this century have shown).

Also, there are various steps that can be taken to lessen the risks of
crypto-anarchy impinging on personal safety.14
Strong crypto provides a technological means of ensuring the practical
freedom to read and write what one wishes to. (Albeit perhaps not in
one’s true name, as the nation-state-democracy will likely still try to con-
trol behavior through majority votes on what can be said, not said, read,
not read, etc.) And of course if speech is free, so are many cla**** of eco-
nomic interaction that are essentially tied to free speech.
A phase change is coming. Virtual communities are in their ascen-
dancy, displacing conventional notions of nationhood. Geographic prox-
imity is no longer as important as it once was.
A lot of work remains. Technical cryptography still hasn’t solved all
problems, the role of reputations (both positive and negative) needs fur-
ther study, and the practical issues surrounding many of these areas have
barely been explored. We will be the colonizers of cyberspace.


-------------------

Notes
My thanks to my colleagues in the Cypherpunks group, all seven hundred of
them, past or present. Well over 100 megabytes of list traffic has passed through
the Cypherpunks mailing list, so there have been a lot of stimulating ideas. But
especially my appreciation goes to Eric Hughes, Sandy Sandfort, Duncan Frissell,
Hal Finney, Perry Metzger, Nick Szabo, John Gilmore, Whit Diffie, Carl Ellison,
Bill Stewart, and Harry Bartholomew. Thanks as well to Robin Hanson, Ted
Kaehler, Keith Henson, Chip Morningstar, Eric Dean Tribble, Mark Miller, Bob
Fleming, Cherie Kushner, Michael Korns, George Gottlieb, Jim Bennett, Dave
Ross, Gayle Pergamit, and especially the late Phil Salin. Finally, thanks for valu-
able discussions—sometimes brief, sometimes long—with Vernor Vinge, David
Friedman, Rudy Rucker, David Chaum, Kevin Kelly, and Steven Levy.
1. RSA Data Security Inc., Redwood Shores, California, is the license adminis-
trator. Contact them for details.
2. Many cryptography texts exist. A good introduction is Bruce Schneier’s
Applied Cryptography (2nd ed.) (New York: Wiley, 1996). This text includes
pointers to many other sources. The annual Crypto Proceedings (Advances in
Cryptology) (Berlin: Springer-Verlag) are essential references. The annual crypto
conference in Santa Barbara and the Eurocrypt and Auscrypt conferences are
where most crypto results are presented.
3. David Chaum, “Untraceable Electronic Mail, Return Addresses, and Digital
Pseudonyms,” Communications of the Association for Computing Machinery 24
(February 2, 1981): 84–88 (cypherpunk-style remailers are a form of Chaum’s
“digital mixes,” albeit far from ideal); David Chaum, “Security without Identifi-
cation: Transaction Systems to Make Big Brother Obsolete,” Communications of
the Association for Computing Machinery 28 (October 10, 1985) (this early
paper is on digital cash; be sure to consult more recent papers).
4. The political opposition in Myan Mar—formerly Burma—is using Pretty
Good Privacy running on DOS laptops in the jungles for communications among
the rebels, according to Phil Zimmermann, author of PGP. This life-and-death
usage underscores the role of crypto.
5. David Friedman, The Machinery of Freedom: Guide to a Radical Capitalism
(2nd ed.) (Ashland, Olt: Open Court, 1989), is leading theoretician of anar-
chocapitalism. Friedrich Hayek was another.
6. Timothy C. May, “The Crypto Anarchist Manifesto,” July 1988, distrib-
uted on the Usenet and on various mailing lists. Also included in this book as
chapter 6.
7. The Cypherpunks group was mainly formed by Eric Hughes, John Gilmore,
and me. It began with physical meetings in the Bay Area and elsewhere and with
virtual meetings on an unmoderated mailing list. The name was provided by
Judith Milhon as a play on the cyberpunk fiction genre and the British spelling
of cipher. The mailing list can be subscribed to by sending the single message,
Crypto Anarchy and Virtual Communities 79
subscribe cypherpunks, in the body of a message to majordomo@toad.com.
Expect at least fifty messages a day. About six hundred subscribers in many
countries are presently on the list. Some are pseudonyms.
8. Abuse, according to some views, of remailers is already occurring. A
Cypherpunks-style remailer was used to post a proprietary hash function of RSA
Data Security, Inc. to the Usenet. Let me hasten to add that it was not a remailer
I operate or have control over.
9. Article on digital cash, The Economist, 26 November 1994, pp. 21–23.
Article on digital cash, Steven Levy, Wired (December 1994).
10. Neal Stephenson, Snow Crash (New York: Bantam, 1995).
11. See Kevin Kelly’s Out of Control: The Rise of Neo-Biological Civilization
(Reading, MA: Addison Wesley, 1994) for a discussion of how central control is
failing and how the modern paradigm is one of market mechanisms, personal
choice, and technological empowerment.
12. During the debate on the digital telephony bill, an FBI official said that fail-
ure to mandate wiretap capabilities within eighteen months would make the bill
moot as the cost would rise beyond any reasonable budget (currently $500 mil-
lion for retrofit costs).
13. “Magic Money” was an experimental implementation of Chaum’s digital
cash system. It was coded by “Pr0duct Cypher,” a pseudonymous member of the
Cypherpunks list. None of us knows his real identity, as he used remailers to
communicate with the list, and digitally signed his posts. Many of us found it too
difficult to use, which is more a measure of the deep issues involved in using dig-
ital analogs (no pun intended) to real, physical money.
14. Robin Hanson and David Friedman have written extensively about scenar-
ios for dealing with the threats of extortionists, would-be assassins, and so on.
Much of their discussion took place in 1992 and 1993, on the Extropians mail-
ing list.