The Changing Shape of Cyberlaw
Lilian Edwards*
Download options |
||
DOI: 10.2966/scrip.010304.363 |
September
is traditionally the time when teachers of cyberlaw across the globe
reconsider the shape of their discipline and how it has changed since
the previous academic year. Cyberlaw is a young discipline –
the first UK case properly so considered was probably Scotland’s
own Shetland Times v Wills1
in 1996, while in the US, perhaps the earliest case to attract wide
attention was the Internet libel case of Cubby v CompuServe2
in 1991 - and one whose minutiae change dramatically from
day to day if not hour to hour. In the struggle to keep up with the
deluge of cyberlaw case law, statutes, regulations, commentaries,
opinions from the technical, economic and political presses,
international treaty activity, European Directives, rounds of
government consultations, conference proceedings and industry Codes
of Practice, it is often difficult for those who observe the domain
to stop and look at the general trends which make cyberlaw in 2004 a
very different animal from when this writer began teaching her first
cyberlaw course in 1997.
But things
are, indeed, very different. In 1997, cyberlaw disputes were mainly
either private law disputes between individual users and other users
or commercial operations, or public (mainly criminal law) disputes
between individuals and states. Many of the latter cases
quintessentially involved states attempting to exert sovereignty over
foreign nationals and corporations whose Internet activities were
impinging on the hallowed enclave of their national laws: in this
vein, the defining criminal cyberlaw case of the 2000s has probably
been the Yahoo! case3,
where French interest groups successfully prosecuted Yahoo! US under
French law for distributing Nazi memorabilia to French citizens, but
then found that the US courts refused to enforce the French decree on
US soil as against US fundamental tenets of free speech4.
The range
of players on the cyberlaw stage has now however expanded. On-line
intermediaries, for example, have always been crucial as gatekeepers
to the Internet and as such frequently both the instigators and the
targets of multiple law suits. But intermediaries now include not
just the traditional Internet Service Providers (ISPs) of this world
– the Demons, AOLs, CompuServes, Freeserves, et al – but
a vast pack of new types of intermediaries whose influence on our
life on-line is profound. No one reading this editorial for example,
is unlikely to get through a day seeking information on the Net
without resorting, thankfully, to Google. Search engines are the sine
qua non of the information society. Aggregators, which combine
information flows from multiple sources on a single page – eg,
giving access to news headlines from the Guardian, the Scotsman and
the New York Times in one place – may well become the primary
conduit to knowledge as the number of on line sources to check out
for current awareness becomes unmanageable. Price comparison
meta-sites – which tell consumers where they can buy the
cheapest travel tickets (say) by scanning 40 sites offering that
service – are a key product to enable consumer choice and
buying power on the Net. “Weblog”5
sites – which enable thousands, sometimes millions, of people
to deliver their thoughts on everything from politics to sex to
recipes to the world – are becoming the new leading mass medium
of communication which need not be filtered (for better or worse)
through the power of an editor or publishing or broadcasting
institution. The different roles these intermediaries play, and the
different policy reasons for supporting them in law, may lead to a
need to fundamentally reconsider existing laws on on-line
intermediary liability.
And, of
course, perhaps the most significant emergence in the intermediary
field of the last five years has been the arrival of the peer to peer
(P2P) sites which allow file swapping and sharing between individuals
– the Napsters, KaZaas, EMules, Groksters and Morpheus’s
of the world. Although best known for (allegedly) facilitating the
illegal downloading of music and films, P2P architectures have the
potential for far more significant abuse – the anonymous and
encrypted sharing of child pornography pictures, for example –
or constructive use – the sharing of information between
scholars, or across states to help the knowledge base of developing
countries. As I write, Kazaa/Grokster has just been found not liable
for contributing to the illegal copying of music files merely by
providing tools which can be so used6.
This decision poses yet again the fundamental question of whether new
types of technology - and the new types of intermediaries which
utilise it – should be encouraged by law in the interests of
innovation, or restrained by law to preserve existing business models
and social norms. The question is a difficult one, more political and
social than legal or moral, and one which will undoubtedly not be
solved any time soon.
The other
new dimension for players on the cyberlaw stage is the international
one. International bodies have, of course, always had a pivotal role
in the development of cyberlaw, given its global nature: intellectual
property law has been driven by the World Intellectual Property
Organisation (WIPO) while UNCITRAL and the OECD have played major
roles in the harmonisation of parts of e-commerce and e-tax law.
Outside of law, the crucial technical standards-setting organisations
that shape the Internet – the Internet Society, the W3C, the
IETF – have always had at least a theoretical internationality
about them even if in practice they were mainly organised by, with
and for Americans. But in the last few years we have seen the
emergence of a truly global, not merely North American,
non-governmental organisation (NGO) sector in the field of
information society policy, concerned both generally with human
rights on line, and with “digital divide” issues in
particular. Here the turning point has been the World Summit on the
Information Society in 2003 in Geneva7
(the next stage being in Tunisia in 2005) which not only proved a
focal point for digital rights activism groups but actually succeeded
in producing a core declaration of principles8
and a tentative action plan9
for the future. What we are seeing here for the digital world is
something akin to the emergence of the UN family of institutions in
the post WWII period: and potentially something of as much importance
for the lives of the ordinary citizen of the developing, if not the
developed, nations.
It is not
just the nature of the actors in cyberlaw that has changed. The
pecking order in importance of the subjects that make up cyberlaw has
too. Once upon a time, IT law was pretty much regarded as all about
intellectual property (IP), really. No one would be foolish enough to
say that IP has ceased to be a crucial part of the cyberlaw syllabus
in these our days of open source, P2P litigation, Creative Commons10,
DRM systems11
and the DMCA12
- but since 9/11 (sadly perhaps) it is rivalled for concern by topics
to do with privacy, security, surveillance and cyber-crime.
E-commerce is another area which has gone from boom to bust to,
perhaps, something approaching the normal expectations of a new
industry sector13,
and the degree of legal attention paid to it has fluctuated
accordingly, although with an inevitable degree - especially in
Europe - of regulation-lag. (The crazy hype and inflated profit
expectations of the dot.com years are now fluently transferring
themselves to the nanotechnology sector. But that is another story
for another editorial.) Whether the amount of law that still pours
out of the European Union to regulate e-commerce can now be justified
(and if it is helpful to an industry already drowning in red tape and
struggling to get past the digital inertia of the average European
consumer) is a question no one seems yet to have asked –
perhaps because in Europe we never did live through the glory days of
dot.com boom quite so fully, and thus the fall, like Icarus, back to
earth has not been so shattering14.
There are
also new topics in the cyberlaw canon: once infrastructure issues
could be covered by a quick scamper through ISP liability; now
cyberlawyers have to know about broadband policy, universal service
requirements, telecoms regulation, European competition law and state
aid, wireless technology and the basics of WTO law. By contrast, some
topics that once looked destined to become hardy perennials of the
curriculum are now looking increasingly cobwebbed and redundant.
Digital signatures, at least as a consumer and small business tool,
have still never really taken hold, despite being the first item on
the legislative agenda of every developing country that decides to
make a show of getting to grips with regulating the Internet. And
digital cash, beloved of banks and merchants seeking to avoid
chargeback losses after credit card fraud, and ignored by consumers
who see no reason to pay by debit in a world of seemingly endless
credit, has essentially died a quiet and unmourned death (just around
the time the EC finally implemented its E-Money Directive15.)
Of course the wheel turns and we are now seeing a new and far more
successful generation of on-line payment mechanisms in the form of
PayPal, SimPay and their ilk – only they don’t seem to
fit into the E-Money Directive after all. The lesson seems to be not
to draft your Directive till your technology has hatched.
So, new
players, new intermediaries, new topics, new priorities in the world
of cyberlaw. But have the forms of regulation themselves changed? Not
at first blush. We still live, as noted above, in a world where every
day it seems there is a new case, a new Act, a new Directive, a new
Convention which in some way attempts to regulate some aspect of the
information society. The European Commission is itself so overwhelmed
by this deluge of legislative activity that it amended its own
Transparency Directive16
so that EU member states come under an obligation to notify the
Commission whenever they promulgate new acts affecting the
information society. But look a little deeper and much has changed.
As is now famously known, in 1999 Lawrence Lessig crystallised the
elegant, not entirely original, but zeitgeist-capturing insight that
in a world of technology, “code” – as in,
primarily, software code – regulates the world as much as
“code” in the old fashioned sense of legal code17.
Since that time, we have seen examples again and again of law being
“trumped” by code: most strikingly in the field of
copyright law, but also elsewhere. In the IP domain, where
traditional law, it at one point seemed, could do little to stop the
Internet becoming “the death of copyright”, “code”,
in the shape of encryption and DRM systems proved able to effectively
prevent illegal copying by anyone but a skilled hacker. (Of course
then code bit right back at the rights-holders, since whatever
copyright work is once unlocked from the chains of DRM can now be
swiftly shared round the world on P2P networks. Code is an
unmanageable beast.) In the privacy domain, the law in the form of
the EC Data Protection Directive aspires to give citizens rights to
control collection and transfer of their personal information without
their consent. But citizens and consumers find these rights almost
impossible to enforce, even if they know they have them. By contrast,
one CCTV camera – hardware “code” – can
breach those rights in a moment, undetectably, and almost
unstoppably. In the ubiquitous surveillance society we now
unknowingly inhabit in the UK18,
code says we have little or no rights to privacy, whatever mere law
may claim.
The
implications of “code as law” for the development of
cyberlaw are interesting. Take the problem of spam, or unsolicited
emails. For the last five years we have seen attempts legally to
control the spread of spam19.
Spam now constitutes around 80% of all email traffic across the
planet and is clogging up the bandwidth of the Internet as
catastrophically as cholesterol blocks the arteries of human beings.
So in the last few years from the US has brought forward the Federal
Can the Spam Act20
and from the EU, we see the Privacy and Electronic Communications
Directive 200221.
Yet the overwhelming likelihood is that the eventual answer to spam
will lie not in law but in code, by altering the nature of the
Internet from an anonymous decentralised system to one which supports
identification, central control and hence a “trusted email”
system: something effectively acknowledged both by the IETF who
control the fundamental architecture of the Internet, and the
International Telecommunications Union at their recent joint WSIS
meeting on spam in Geneva22.
This writer, who (for her sins) is currently writing a chapter on
the legal regulation of spam, strongly suspects that in perhaps five
to ten years’ time this will be a solved technical problem and
no longer a legal concern at all. If this applies to other areas of
cyberlaw as it does to spam, then we might expect to see cyberlaw
begin to contract rather than exponentially expand as it has done so
far. Yet if we so imagine we may also be falling foul of a sort of
technology-supremacism which fails to fully appreciate the social as
well as the technical dimension to many cyberlaw problems. What is
essential is for legislators and judges to make sure that the future
products of their labours are not “trumped” by code but
work hand in hand with it.
And of course, “code” as law has engendered a new breed of law, law which in its turn attempts to tame code. So far, the most famous example of this breed is the US Digital Millennium Copyright Act, and its attempts to make it a crime to write or distribute code which hacks technological copy-protection mechanisms imposed by rights-holders. But there have been other examples of law which (tries to) regulate code: Lessig himself refers back in his early essay on “The Law of the Horse” to the US government’s vain efforts in the 1990s to limit access to “strong” encryption via means such as imposing the “Clipper Chip” on encryption users23. In general, law which tries to trump code seems somewhat less successful than code which effortlessly trumps law. But it is all good news for lawyers and legislators, who need never fear an absence of work while this digital version of la ronde continues.
* Co-Director, ahrc Centre for IP and Technology Law, Edinburgh
2 776 F.Supp 135 (SDNY 1991).
3 TGI Paris, Ordonnance de refere du 20 Nov 2000. This order confirmed the earlier decision issued on May 22 2000.
4 Yahoo! v La Ligue Contre Le Racisme et L’Antisemitisme 169 F. Supp. 2d 1181 (N.D.Cal 2001). But see also now http://www.law.com/jsp/article.jsp?id=1090180400752 .
5 See eg, Blogger, Moveable Type, LiveJournal.
6 MGM v Grokster, appeal to US Court of Appeal for 9th Circuit on August 19 2004. See http://www.eff.org/IP/P2P/MGM_v_Grokster/20040819_mgm_v_grokster_decision.pdf .
10 http://creativecommons.org/. This editorial is not in fact, as you might perhaps expect, licensed under a Creative Commons license but under SCRIPT-ed’s own open access license – see http://www.law.ed.ac.uk/ahrc/script-ed/sol.htm
11 Digital Rights Management.
12 Digital Millennium Copyright Act.
13 See The Economist “Paradise Lost” May 8 2003 at http://www.economist.com/surveys/displaystory.cfm?story_id=1747329
14 See further Edwards L ed The New Legal Framework for E-Commerce in Europe, forthcoming, Hart Publishing, February 2005.
15 Directive 2000/46/EC of the European Parliament and of the Council of 18 September 2000 on the taking up, pursuit of and prudential supervision of the business of electronic money institution.
16 Directive 98/34/EC as amended by Directive 89/48/EC/.
17 Lawrence Lessig, Code and Other Laws of Cyberspace (Basic Books, 1999).
18 See further Goold B J CCTV and Policing (Clarendon Studies in Criminology, OUP< 2004).
19 See this writer’s early summary (for Europe/UK) of legal regulation of spam as of the year 2000: Edwards L “Canning The Spam” in Edwards L and Waelde C eds Law and the Internet: A Framework for Electronic Commerce (Hart, 2000).
20 Which came into force on 1 January 2004.
21 2002/58/EC.
22 See ITU WSIS Thematic Meeting on Countering Spam, Geneva, 7-9 July 2004.
23 See Lessig L “The Law of the Horse: What Cyberlaw Might Teach” (1999) 113 Harvard Law Review 501 at 532ff; available at http://www.lessig.org/content/articles/works/finalhls.pdf. Lessig rather cutely describes this episode as an attempt by “East Coast Code” (Washington legislators) to compete with “West Coast Code” (encryption software writers in Silicon Valley).