Skip to main content

JILT 2003 (1) - Konstantinos Komaitis






ICANN: The Legal Status


ICANN is Engaged in Policy-Making


DoC's Relationship with ICANN


ICANN is Violating Basic Competition Laws


The Effect of ICANN's Actions in Europe




Notes and References

word icon and download article in .rtf format Download

ICANN: Guilty as Charged?


Konstantinos Komaitis
University of Strathclyde

Keywords: ICANN, NSI, DNS, DoC, UDRP, Competition Law.

This is a Refereed article published on 4 July 2003.

Citation: Komaitis, K, 'ICANN: Guilty as Charged?', 2003 (1) The Journal of Information, Law and Technology (JILT). <>. New Citation (as at 1/1/04): <>.


1. Introduction

'ICANN has been given too many breaks - but ICANN's hubris dates from before its inception. ICANN's disdain for the Internet user is intense. ICANN is trying to be the Louis XIV of the Internet, but with less skill', Karl Auerbach.

Let's consider the acronym ICANN. It stands for Internet Corporation for Assigned Names and Numbers. Who is entitled to assign 'Names' and 'Numbers' is basically in the position to control the Internet entirely. Indeed, the whole structure and foundation of the Internet depends on names and numbers.

ICANN is formally a private non-profit California corporation created, in response to a summoning by US officials, to take the necessary regulatory action that the Department of Commerce (DoC) was unable or unwilling to take directly. The corporation has been strongly criticised, mostly accused than praised, but at this point in time it is the one and only body that is officially responsible for the management of the Internet or - in other more technical words - of the domain name system (DNS).

Despite being famously decentralised and un-hierarchical, the Internet lies on an underlying centralised hierarchy built into the domain name system[ 1]. Domain names - such as are the unique identifiers that people depend on to route e-mail, find web sites, and connect to other Internet resources. The need to enforce uniqueness, that is to prevent two people from attempting to use exactly the same domain, creates the need for some sort of body to monitor and allocate naming. However, control of the DNS equates to substantial power over the Internet. Whoever controls the DNS decides what new suffixes will be added to the root server and how names and essential routing numbers will be assigned to websites and other Internet resources[2]. The power to create is also the power to destroy, and the power to destroy carries with it the power to attach conditions to the use of a domain name[3]. Currently, this power is used to require domain name registrars to publish their addresses and their telephone numbers on a worldwide readable database - namely the notorious WHOIS- and to agree that any trademark holder in the world aggrieved by their registration can demand arbitration regarding ownership of the name under an eccentric set of rules and numbers[4].

Without meaning to at first, and without realising the power that it held in its hands, the United States government found itself in control of this exceptional checkpoint. However, as the DNS issue and in particular the relationship between trademarks and domain names grew significantly, the pressures on the federal government for action grew as well. In June 1998, the DoC and an interagency task force headed by Presidential Senior Adviser Ira Magaziner responded with the Statement of Policy on the Privatisation of Internet Domain Name System, known as the DNS White Paper[5]. Abandoning earlier hopes of issuing a substantive rule, which requires statutory authorisation and is subject to judicial review, the policy statement instead set out objectives that the administration planned to be achieved without rulemaking. Using the rhetoric word 'privatisation', the DNS White Paper talked about the creation of a private, technical and non-profit corporation to take over the management and put forward various plans for reform of the structure and administration of the DNS[6]. Shortly thereafter, ICANN spurred out of the premises of the White Paper as a California based corporation and soon after some negotiations and debates, the DoC 'lent' ICANN most of the authority and responsibility that comes with the supervision of the DNS[7].

ICANN's creation drew interest from a diverse group of stakeholders, including Internet users, domain name registrars, technical groups and intellectual property law associations. While each group offered different perspectives on issues such as domain name dispute resolution and the creation of new suffixes, there was widespread agreement on one key principle - ICANN was to be based on a self-regulatory model free from government interference. ICANN never really upheld its end of the self-regulatory bargain. While the stakeholder groups assumed their guaranteed positions on the ICANN board, only five of the nine elected seats were ever filled because the appointed board never put the remaining four seats up for election, thereby creating an imbalance that has never been rectified[8].

As soon as it was organised, the ICANN board undertook major decisions, beginning with the agenda set out in the White Paper. ICANN pushed Network Solutions, Inc. (NSI), the monopoly registry and sole registrar, to allow more competition among registrars[9]. ICANN also introduced mandatory arbitration of trademark claims. ICANN's 'Uniform Domain Name Dispute Resolution Policy' (UDRP) requires every registrant in .com, .net and .org to agree to arbitration before ICANN- accredited arbitration providers if any dispute among trademark owners and domain name holders occurs[10].

All these manoeuvres, although out of the initial scope of ICANN, were the cause of a lot of concern in the Internet community. ICANN was gaining ground and power with the 'silent' approval of the DoC. There is substantial evidence, discussed below, that the DoC has directly instructed ICANN on policy matters. Furthermore, since ICANN is utterly dependant on the DoC for ICANN's continuing authority, funding and thus its existence, it would be reasonable to conclude that the corporation is currently so captive that most - if not all - of ICANN's decisions can be charged to the government[11].

Moreover and in respect of internal relations, ICANN falls short of much of the accountability normally found in corporations and in not-for-profits. Ordinary corporations have shareholders and competitors. ICANN does not, because it is non-profit and has a bizarre relationship with the DoC. Many non-profit organisations have members who challenge corporate misbehaviour. ICANN has taken steps to ensure that its 'members' are denied such legal redress under Californian law. All but the wealthiest non-profits are constrained by needing to find financial resources. ICANN faced such problems in its early days, but it has now managed to overcome them by affirming its power over the legacy root, accepting contributions from the registrars that have agreed to recognise ICANN's authority for the .com, .net and .org registrations and from NSI, the dominant registrar of the three commercial TLD's, which agreed to pay a substantial amount of money to ICANN as part of agreements reached with the DoC and ICANN[12].

However, the US Government did not plan to be in charge of the Internet. Its control over the root was, more than anything else accidental. Nevertheless, as soon as the Clinton Administration realised its influential and political power, it welcomed its authority over the root and did not really attempt to divert its powers. Thus, like the story of Sherlock Holmes's dog that did not bark in the night, the ICANN story contains a telling absence. ICANN lacks statutory background. There is no evidence, among the Memoranda of Understanding, the contracts or the bilateral agreements that ICANN has signed throughout the years, that Congress has ever authorised ICANN or the 'privatisation' of the DNS[13].

ICANN is currently going through one of the most difficult periods since its existence. What used to be rumours and speculations seem now to be facts and some are willing to take action against the corporation[14]. It is rather easy to impugn the role and integrity of ICANN; nevertheless, such an action does not really facilitate nor does it help to give solutions. This paper will not escape this normality, but it will try to point out that perhaps ICANN's actions are not autonomously; rather they are directed by the US Department of Commerce. This paper will demonstrate how ICANN is engaged in policy-making and how its relationship with the DoC is of a strange nature. Finally, it will illustrate that the corporation is operating in a way that distorts competition and engages ICANN in anti-competitive practices.

1.1 ICANN: The Legal Status

Is ICANN a technical or a policy-making body? Is ICANN a private or a government corporation? Is ICANN breaching any laws? These are three questions of vital importance concerning the legal status of ICANN. If ICANN is a policy-making body, then it violates its agreements and its MoU with the US Government. If the DoC controls ICANN, ICANN is not, as it declares, a private not-for-profit organisation and it infringes the agreements and the MoU with the DoC. If ICANN's legal status is contrary to the existing laws then ICANN should be liable.

This part begins by arguing that at least some of ICANN's activities pursuant to its agreements with the DoC are governmental in nature, and can be classified as policymaking or regulation. The subsequent sections examine ICANN's relationship with the DoC in light of the contractual history and will mention the most significant breach of law that takes place; breach of the US antitrust law in parallel with EU competition law.

1.2 ICANN is Engaged in Policy-Making

Both ICANN and the DoC proclaim that ICANN is not engaged in policy-making matters. Instead, they offer the standard-setting view, according to which ICANN's role is nothing more than routine standard setting or perhaps technical coordination[15]. Thus, ICANN maintains that it 'has no statutory or regulatory 'authority' of any kind'[16]. It has, it advocates, 'only the power of the consensus that it represents, 'since root servers not run by or under contract to the US government could at any time choose to point to a different root, albeit at the cost of causing the very fragmentation most root server operators oppose'[17]. ICANN argues that only 'the willingness of members of the Internet Community to participate in and abide by the consensus development process that is at the heart of ICANN'[18] makes people do what it says, not coercion or regulation[19].

ICANN's proper role consists of two jobs and two jobs only:

  1. Making sure that IP addresses are assigned and allocated on a fair and equitable basis and in conformity with demands of the packet routing systems of the Internet.

  2. Making sure that the ICANN/NTIA root zone is expanded on a basis that is fair and equitable to everyone, that the root zone file is properly maintained and disseminated, and that its set of root servers are operated by persons and entities that have the proper skills, resources, and obligations[20].

However, choosing new suffixes to be added to the root zone on the basis of social utility from among multiple technically qualified providers, fixing the business models of registrars, enforcing dispute resolution procedures on millions of unwilling businesses and consumers, accrediting dispute resolution providers, writing substantive and procedural rules for the disputes, none of these tasks are 'technical' standard setting in any routinely meaningful sense of the world[21].

The most shining example of ICANN's policy-making to date is beyond any doubt its promulgation of the Uniform Domain Name Dispute Resolution Policy (UDRP), following the lead set out in the White Paper. The White Paper identified cybersquatting as a major problem that would need to be approached by NewCo, and called for the UN's World Intellectual Property Organisation (WIPO) to provide NewCo with an advisory report on how the problem might be best approached[22]. WIPO 'gladly' produced a lengthy report[23]. ICANN referred the question of arbitration of cybersquatting cases to a 'working group'. While ICANN's working groups are expected to be open to all, in this particular one the Chair excluded several opponents of mandatory arbitration from voting, although everyone was allowed to submit comments[24]. The Working Group's report was then forwarded to the ICANN Names Council, whose role is to assess whether there is consent for the report. It did not, instead choose to air its own views on the matter[25]. After a period of public comment, the ICANN board then took up the issue. Finding that there were some unresolved and complicated issues, the board created an ad hoc committee and dubbed the 'small drafting committee', that met in secret to advise the staff on recommendations to the proposed policy, some but by no means all of which the staff accepted[26]. The board then adopted the staff's final proposals.

By adopting these rules, ICANN imposed on all current and future registrars in .com, .net and .org domains a requirement that they agree to a third-party beneficiary clause in favour of any person, anywhere in the globe, who believes that the registrant's domain name registration infringed the complainant's trademark right. The clause though was not one of an option; pursuant to the Registrar Agreement, ICANN required domain name registrars to include this clause in every registration contract and to modify existing contracts with their customers; and, pursuant to the same aforementioned agreement ICANN required NSI to agree to refuse to list registrations from any registrar who failed to do so[27].

The DoC's, WIPO's and ICANN's objective in promoting a mandatory domain name dispute resolution was to create a fast, inexpensive, lightweight process that would allow victims of cybersquatting to vindicate their rights easier than would be possible in most courts[28]. In a way it indeed has, but it was not in any sense 'technical coordination' of the Internet. Rather, it represents a clear policy choice to sacrifice the interests of domain name registrants in favour of trademark registrants for the communal good. While, the policy is one, which a legislator could draw, there is not evidence that Congress has delegated power over trademark policy to the DoC[29].

One of the most worrying aspects of the UDRP is that ICANN's initiatives on devising the Policy have generated a web of issues concerning the nature and efficiency of the system. No one really expected from ICANN to bear the responsibility of formulating a dispute resolution Policy, taking into account the fact that ICANN's creation was based on a technical charter and this was a policy making decision. ICANN, nevertheless, decided to have the final saying on the way the system would operate and more or less foreshadowed WIPO's proposals; thus, ICANN produced the UDRP, but its task ended there. ICANN officially declines any sort of responsibility of the way the Policy is interpreted or the disputes are approached and decided; so do the four accredited by ICANN arbitration centres. However, ICANN should be held responsible for the decisions, because it is ICANN in the first place that created the Policy and it the only authority, which can amend any of its rules. Therefore, taking into account that ICANN is willing to engage itself in policy-making decisions, then it should be held liable for the way the Policy is progressing[30].

Consequently, the standard-setting's defence that this is a mere technical coordination is not credible. Technical issues involve questions such as the bit size of data packets, the architecture of the root servers, or the number of new top-level domain that can safely be added to the root. Determining what sort of redundancy the DNS requires and investigating the Y2K liability of the DNS are also technical questions. Changing the legal rights reflecting more than twenty millions of registrations is not[31].

Even though the clearest case of all is the adoption of the UDRP, this is not a unique case. As it prepares to add more TLDs to the root, ICANN will once again venture outside of the technical arena[32]. In the meantime, ICANN chose the seven new suffixes - the 'magnificent seven', as they ended up being known, - from around two hundred applications from potential registries who paid a non-refundable fee of $50,000. Some proposals were rejected for not demonstrating sufficient technical abilities. Taking into consideration that there were applicants with more technical skills than ICANN was willing to grant a TLD, then ICANN chose the applicants according to other criteria. Specifying minimum performance levels for new registries and registrars are indeed technical matters, but choosing among multiple competing, technically qualified, proposals to see which of them suits ICANN better based on the social value of the proposed name does not really constitute 'technical coordination'[33].

Finally, even though ICANN does not proclaim itself as a regulatory body, it has in fact engaged in a process of regulation by contract, which has resulted in a string of substantive policy making. This has not been achieved by the issuance of 'rules', but rather through the drafting of private contracts. For instance, when ICANN renegotiated its contract with VeriSign, it argued that altering the terms of its contractual relationship with VeriSign would enhance competition in the DNS marketplace. That may be a desirable goal for ICANN and a step forward, but making rules to promote competition is a task that governments and not private entities undertake[34].

The claim that ICANN is exercising governmental regulatory functions in not new. The EU Commission noted that '[e]ven within their narrowly defined remit, it is already the case that ICANN and the [Governmental Advisory Committee] are taking decisions of a kind that governments would, in other contexts, expect to take themselves in the framework of international organisations'[35]. Not only is ICANN making regulatory decisions for the whole DNS, but the structure of the contracts set up by the DoC make ICANN into the regulator of the registries and even the registrars. In particular, ICANN is now NSI's regulator; in particular amendment 19 to the DoC-NSI agreement states that NSI recognises ICANN as NewCo[36].

There is no doubt that ICANN is engaged in 'policy' - the establishment of a supranational law of trademarks versus other rights, a law called the UDRP, is no small matter and amounts to an usurpation of legislative rights of the nations of the world. However, making policy is not illegal - there is the old and obsolete statement that 'nature abhors a vacuum'. There is a vacuum with regard to the establishment of the proper relationship between domain names and other uses of names. ICANN's actions are being directed by the efforts of the US DoC to satisfy the wishes and desires of trademark holders; wishes and desires well beyond that which any national legislature was willing to provide[37].

1.3 DoC's Relationship with ICANN

The DoC's relationship with ICANN is defined by two sets of opposite claims that are hard, perhaps legally impossible, to reconcile. On the one hand, it is argued that the DoC retains ultimate control over the root zone file; on the other, it committed itself to the 'privatisation' of the DNS and since then its actions are consistent with the effort to being seen to control the DNS[38]. Until ICANN came along, the DoC undoubtedly had the sole control of the root and the DNS. Some examples include the original policy against registering multiple domain names, commercial uses, and a number of NSF policies directed to encourage use on the Internet, including making registration easy. Although ICANN claims that all domain name registrars and registries are subservient to it, in reality the DoC enforced this subservience. In more practical terms, the DoC's authority over ICANN and the root rests in three elements: its agreements with NSI, its ability to hold a sword of Damocles over ICANN and its direct teamwork with ICANN[39].

It is rather difficult to ascertain whether the DoC has the authority to transfer control of the authoritative root server to ICANN. Even though control over the 'A' root server is not based on any international or statute agreement, the US Government has always been influential with the administration of the Internet and the Domain Name System (DNS). Federal statutes in the United States do not incorporate provisions obliging the Government to manage the DNS or to control the authoritative root. Similarly, it is tentative whether transfer of the control of the root would equate to transfer of government property to a private entity; however, determining whether there is government property in the first place is also difficult. Taking into account though the fact that the whole function of the Internet and the DNS is based upon the very function and control of the 'A' root server, by creating ICANN, the DoC would have had two options; to transfer the control of the root to the corporation or not go through this process at all. Nevertheless, it is uncertain whether such a transfer is either permissible or acceptable. The DoC has clearly stated that it has no plans to transfer the control of the root system to either ICANN or anyone else. Currently, under the cooperative agreement with Network Solutions, the DoC has reserved final policy control over the authoritative root server[40].

One of the various wonders of the privatisation of the Internet and the DoC's involvement is whether the process of the privatisation and the subsequent establishment of ICANN were subject to the Government Corporation Control Act. The Act provides that '[a]n agency may establish or acquire a corporation to act as an agency only by or under a law of the United States specifically authorising the action'[41]. Although the Department's Policy statement contemplated the creation of a private not-for-profit entity, in reality the DoC did not establish or acquire ICANN as such. ICANN was created from the initiatives of Jon Postel and his associates, who sought to transform IANA into a corporation compatible with the demands of the privatisation process. Therefore, the Government Corporation Control Act does not apply per se[42].

However, it can be argued that the Act may be applicable, albeit in an indirect way, which still proves that the DoC is perhaps violating the Act's basic concept. It is beyond any reasonable doubt that the DoC's role and influence on ICANN is sustainable and important; in case the DoC was not satisfied with the way Jon Postel decided to structure ICANN, it would not authorise its creation. Even though the DoC did not interfere in the drafting of ICANN's bylaws, still it accepted and took the responsibility that ICANN is indeed capable of carrying out the management of the DNS. The DoC had the final saying of how the corporation would look like and what would its job be and it used the White Paper to enforce its aspirations. Therefore, whether or not the DoC established or acquired ICANN directly is not that important in this particular case; what is important is that the DoC gave its silent approval to Jon Postel and his team to create a corporation, in which the DoC would have an active role in at least some significant decisions.

Moreover, the relationship between the DoC and ICANN depends on the various Memoranda of Understanding (MoU) that the two parties have signed. In the MoUs it was clearly stated that ICANN would take up the job of privatising the Internet, a task that IANA was responsible for. While the MoUs are not binding by nature, however they ended up having a great influence and they are considered to be the pillars of ICANN's existence. Thus, it is viable to argue, albeit quite extreme, that a MoU is a form of quasi-contract and thus falls within the Government Corporation Control Act. However extreme this approach may be, it is actually a fact that the DoC is involved in the management of the DNS and in ICANN one way or another, which is actually the pivotal conclusion.

What is more, despite all the agreements among ICANN, the DoC and NSI, ICANN is not able to approve new suffixes without DoC's approval. Even though NSI is the physical authority, which controls the root, it remains obliged to secure written approval from the DoC before adding new TLDs to the root[43].

Similarly, ICANN has always to make decisions according to the DoC's requirements, as stated in the White Paper. Indeed, if, for any reason, ICANN fails to meet the DoC's expectations, and the latter recognises another authority as NewCo, then ICANN's days are over[44]. Furthermore, the DoC and ICANN work on a cooperative basis, although it is claimed that their relationship is equivalent to a partnership. Yet, no matter how is labelled, the fact is that it was obvious from the beginning that the DoC planned to be as involved as possible with ICANN. Although, the DoC did not sponsor ICANN during its financial crisis, it instructed NSI to 'donate' to ICANN a substantial amount of money and at the same time it decided to spend around a quarter of a million dollars in staff time and expenses to monitoring and helping ICANN with its tasks[45].

In the same vein, any decision of ICANN, except those that have truly a technical character, is subject to the DoC's previous approval. If this is the case, all ICANN's policy decisions take effect, just because the DoC passively supports them. The relationship between ICANN and the DoC may be of a strange nature, but whether the latter actually and directly controls the corporation, is namely a speculation. What is not a speculation however, is that although the DoC can exercise its control, nevertheless it seems as if it interferes occasionally and whenever it deems necessary to do so.

It is somewhat a mistake to claim that generally the Doc has absolute control. ICANN, being a private body, is legally beyond the reach of the DoC except via the agreements that the DoC has with ICANN. Those contracts are extremely powerful - if ICANN were to lose them, its funding would terminate. However, the US government has to act with 'due process' and may not be arbitrary and capricious. Nor may the US government interfere into private matters until and unless there has been a properly enacted law permitting some designated agency of the government to do so. ICANN is in many respects more subject to regulation by California - there are for example things called 'police matters'; and in the United States those powers belong to the individual states[46].

All these speculations though seem to lead to one concluding result; the DoC is interfering with the way ICANN is performing its functions, either directly or indirectly. Realistically speaking this makes a lot of sense. Conceptually, the Internet goes beyond the communication skills that other means of information have to offer. The Internet has discarded all the existing barriers and has allowed information and communication to flow in a way that no one has predicted or anticipated. Therefore, control of the Internet supports the dynamics of governance and equates to substantial economic and political power. In the beginning the US Government was unwilling to be in charge of the medium, merely because they were not aware of its dimensions; however, now they seek to be part of its management one way or another. Perhaps the US Government is willing to take back the control they once had and this is further supported by the recent bill proposal by the US Senator Conrad Burns, who suggested that the US government should have more influence over ICANN than it already has so as to ensure appropriate scrutiny. Unfortunately, such an approach would only make things worse, with ICANN already coming under fire for being too US-dominated. If the US government were to exert more control '[i]t could be seen as an effective annexation of the internet [sic]'[47]. The bottom line thus remains that ICANN's actions are not so self-directed as many want to believe and thus criticising solely the corporation is of no point; it would be presumptuous to speculate that the US Government is envisioning to dominate the Internet the way it used to some decades ago.

1.4 ICANN is Violating Basic Competition Laws

Recently a new accusation against ICANN occurred, which added concern to ICANN's existence. It was the possibility of ICANN violating US antitrust laws.

One of the most fundamental and revolutionary principles of antitrust laws in the United States is that government agencies and those they authorise to act are immune from antitrust scrutiny. The source of such immunity is the 'state action doctrine', which strictly speaking protects actions by states and those they represent, perhaps including municipal governments as well[48]. The general rule is that in order to grant immunity the government must not only have the right to overrule private decisions, but also to be able and review such decisions[49]. The principle is fundamental because it enables the government to determine the operation of the competitive market via its agencies and to set the standards for such an action.

Prior to ICANN, NSI was the sole registrar and administrator of the domain name system. Despite the rather well designed framework for antitrust immunity, the Doc found itself in a quandary, due to the fact that many of the NSI's actions could not be protected under the fibre of the immunity doctrine. Many reported cases have considered antitrust claims against NSI's anticompetitive conduct towards the management of the DNS and its dispute resolution policy; and, even though district courts were prudent and refrained from imposing sanctions for the conducts alike, the appellate courts were more cautious in granting such immunity[50]. Needless to say that one of the main reasons that the DoC was not willing to give NSI the role that ICANN currently has was because of the lawsuits that have taken place; it was thus considered that NSI would not be liable enough to undertake the task of administering the Internet. Nevertheless, it looks as if ICANN is involved in a string of allegations concerning violation of antitrust laws, but as of yet no court action has been taken.

During the four years of its existence, ICANN made important progress on several of its assigned tasks related to promoting competition. At the time the transition began, only NSI was authorised to register names. In response to the MoU task calling for increased competition, ICANN successfully developed and implemented procedures under which other registrars could carry out this function.[51]. As a result, by early 2001 more than 180 registrars were certified by ICANN. The cost of securing these names has now dropped from $50 to $10 or less per year. Moreover, ICANN was called to add new TLDs; to test the feasibility of this idea, ICANN's board selected seven new suffixes and by March 2002, it had approved agreements with all seven of the organisations chosen to manage the new domains[52]. Whilst ICANN has promoted competition, at the same time some of its actions prove that it has restricted it and thus it infringes section two of the Sherman Act, which states 'every person who shall monopolise, or attempt to monopolise', a relevant market[53].

ICANN's decision to limit the number of new gTLDs created an artificial scarcity of domain names. It has also limited the number of companies who could be registries, since the DNS - the way we know it - presupposes that there will be only one registry for each TLD. ICANN's method of choosing registries presents rather serious antitrust issues[54]. ICANN's application document for would-be new TLD registries, and especially the accompanying[55] 'Criteria for Assessing TLD Proposals', made it clear that parties who dealt with ICANN's competitors would be rejected, although those competitors' market share was significant[56-57].

Moreover, ICANN unquestionably has control - along with the DoC - over the legacy root. Virtually all gTLDs (measured by use) are under ICANN's effective control, and ICANN's control over the root creates substantial obstacles to entry to alternate roots[58-59]. To be more precise, ICANN has total control over ccTLDs because it controls the root to which they are linked. Even though the political cost for ICANN would be enormous, if it decided to use its power over ccTLDs, there are things that ICANN can do to ccTLDs, such as removing them from the root. For example, although ICANN has a contract with the US government to perform the functions of IANA of maintaining the root, including ccTLD information, ICANN has a policy of refusing to make any changes to the contact information and data to any ccTLD that has not been contracted with it[60].

Nevertheless, ICANN's exclusionary conduct towards competitors is exemplified in its treatment with ICANN made it clear from the beginning that potential customers should understand that their registrations will never be recognised in the ICANN root. First, ICANN's CEO attacked of 'breaking the Internet' and 'selling snake oil', while he authored a paper questioning at the same time the company's legitimacy that he labelled a 'discussion draft'[61-62]. Then without any warning, ICANN announced that a slightly revised version of the paper was official ICANN policy, and that no 'bottom-up' discussions were needed, as the paper restated long-standing policy rather than suggesting something new[63]. ICANN's opposition to is derived from the fact that the company has managed to capture some rather appealing domain names and add them to an alternate root, which has secured through agreements with various ISPs. An example, of ICANN's conflict with happened in connection with ICANN's Annual Meeting in November 2001. In response to ICANN's general call for sponsorship, sent in a $5,000 fee, which would have entitled it to place its logo at a coffee break during the meeting, and to distribute company material at the sponsor's table. ICANN returned the cheque, stating that was not welcomed since it did not support the authoritative root[64].

Moreover, ICANN's role in selecting companies to become approved registrars has also come under fire. On the 26th of June, 2000, ICANN was sued by, a New York company that ran a website where people who owned domain names could resell them. had applied to become an accredited registrar, but its application was rejected without further explanation. Since ICANN is the only authority that can give authorisation to a registrar to register domain names in .com, .org and .net, ICANN's denial automatically meant that would be unable to resell domain names. The suit was settled on the 7th of July, 2000, but it may well be a sign of hazardous times for ICANN[65].

Furthermore, another example of ICANN's anti-competitive status is the UDRP. ICANN required all registrars to agree to impose a mandatory dispute resolution on their registrants. By doing so, ICANN entered into a vertical agreement restricting non-price competition on one axis. The Policy was adopted to limit cybersquatting behaviour and to serve the smooth existence of domain names and trademarks. There is strong evidence though that the UDRP was truly enacted at the behest of intellectual property owners who likely had the political power to prevent the adoption of any new gTLDs unless the registrars agreed to restrict cybersquatters[66]. The effect of the UDRP thus becomes clear; eliminating competition that would have otherwise existed between registrars about how to solve disputes. Some claim that it is a mistake to argue and rather difficult to establish that the UDRP proves an anti-competitive process. The UDRP is not however against competition per se; it is contrary to regulatory competition, a form of competitive conduct, which has the results of pure competition as it does not give any room for choice of action and obliges the parties to abide to only one set of rules.

Considering all the aforementioned issues one of the questions that emerges is whether ICANN falls within the principles for antitrust immunity. If ICANN has a relationship with the DoC, then it falls within the meaning of antitrust immunity; if, on the other hand, the corporation is indeed a private entity, then the principles of antitrust immunity will not be applicable. What is the case and is antitrust immunity applicable in the case of ICANN?

Given the strange relationship between ICANN and the DoC there is not a clear answer. Initially it needs to be stated that ICANN will not be able to rely on absolute form of federal immunity, since Congress has not created an expressed at least exception to the antitrust laws for ICANN. Thus, if ICANN is to be immune it would most probably depend on its contractual relationship with the DoC; however, appellate courts have not gone far enough to endorse the theory that any government contractor is entitled to absolute immunity. In the case of ICANN the best option would be to take each case separately and decide whether ICANN's actions are immune or not. The matter seems rather hazardous. Officially both ICANN and the DoC state that there is no relationship between them, and ICANN's actions are not directed by the DoC. Of course it is always possible that ICANN would be able to demonstrate that the DoC has a behind-the-scenes interference in ICANN's decisions, but it would be difficult perhaps even for ICANN to sustain such an argument[67].

The White Paper can be used to argue both scenarios of the 'clearly articulated government policy' test. On the one hand, the White Paper itself considered and rejected the idea of 'NewCo' given antitrust immunity. Indeed, in the White Paper it is stated '[a]pplicable antitrust law will provide accountability to and protection for the international Internet community. Legal challenges and lawsuits can be expected within the normal course of business for any enterprise and the new corporation should anticipate this reality'[68]. This seems as a rather articulated policy that ICANN cannot be subject to antitrust immunity.

On the other hand, the White Paper contained a number of Policy directions and instructions that the NewCo should take on board and ICANN has almost literally followed those instructions. Amongst other things, the White Paper for example, instructed ICANN to devise a dispute resolution policy for the conflicts between trademarks and domain names. The ultra vires argument rests on the fact that since the DoC lacked any statutory authority to devise such a system, it authorised ICANN to do so on its behalf. Subsequently, the thighs between ICANN and the DoC seem to be strong and thus ICANN is subject to antitrust immunity, even if the White Paper can be used to prove this theory[69].

What the truth is behind ICANN is rather ambiguous and a matter of debate. However, it can easily be argued that some of ICANN's actions are violating antitrust laws in the United States and occasionally the violation is supported by the silent approval of the DoC. Nonetheless, until ICANN finds itself involved in antitrust lawsuits and until the court decides on the matter, these are just speculations.

1.5 The Effect of ICANN's Actions in Europe

Moving on the other side of the Atlantic, an analogy can be drawn to attest that ICANN is also violating basic principles of EU competition law, in particular Articles 81 and 82 of the EC Treaty. ICANN is not an international organisation, as it was not created by a Treaty; however, it is accepted that ICANN is a quasi organisation with international consensus and the main responsibility of allocating names and numbers on the Internet. It is a corporation that deals with an issue, which has extraterritorial effects, thus the allegations for ICANN's anti-competitive practices that seem to aggravate the US commentators have found their application within the European Union as well.

Competition law is the cornerstone for the European Union's political, social and economic framework; its enforcement is one of the EU's success stories, because of the strong principles that surround competition policy. The European Commission has soon realised that the essence of competition is to create a multi-faceted market, where stimulation and innovation would be able to flourish; for these reasons it has identified three focal points that are able to distort competition and constitute the core of competition policy within Europe. These are the following:

  • Market dominance; and

  • The question of specific threats to competition caused by B2B platforms.

Even though the three principles incorporate in their meaning the rationale of competition, their application is not always successful. In particular, when it comes to ICANN, it is rather difficult to prove anti-competitive practices - at least the way Europe understands them - only because ICANN is lacking one important element; ICANN is not abusing its dominant position in either the product or the geographical market. According to article 82 of the EC Treaty, an undertaking can only be abusing a dominant position in relation to a particular market and can mainly consist of two different markets, namely the product and the geographical one[70]. ICANN though is not providing products that are able to dominate a geographical market; the corporation is responsible for the administration and organisation of the domain name system and is only providing services. Therefore, how possible is it for a corporation to eliminate competition by providing services?

The answer is rather vigorous; the same rules that apply in the case of product domination, apply when an entity is abusing its market domination and provides services. All of ICANN's actions prove anti-competitive practices and are contrary to EC Competition rules. Everything of what it seems to be speculations under the US Antitrust laws, are a fact within the European Union. ICANN is in a dominant position and is abusing this dominance willingly and consciously. The fact that the corporation is able to determine whether Internet users will make it to the root, how long they will stay and whether they will be visible or not, is enough evidence that ICANN is in a dominant position. What is more, the possibility that ICANN shares to deny to list both gTLDs and ccTLDs to the 'A' root zone file is just adequate to demonstrate that the corporation is abusing its dominance in the market. However and while it should be taken for granted that ICANN will never deny to list a ccTLD to the root, as the political implications would be radical, the mere fact that ICANN has the option to determine who will make it to the Root is sufficient to attest that under European competition laws ICANN is engaged in anti-competitive techniques.

The Microsoft case can be exemplified, in order to understand the way ICANN, a US based corporation, is violating competition laws within the European Union. Microsoft was brought before the US courts for violating federal and state antitrust laws in the United States based on the allegation that users, once the Windows operating system was installed, they were given the option of using Internet Explorer and not Netscape Navigator. The United States court concluded that Microsoft maintained its monopoly power by anti-competitive means and attempted to monopolise the Web browser market[71]. However, in Europe things turned out to be more difficult for Microsoft, since the European Commission decided also to forbid Microsoft from merging with other companies, which would automatically declare the company as the dominator of the software market; this would most probably equate to abuse of Microsoft's dominant position within Europe. Microsoft was being investigated for violating Article 82 of the European Community Treaty, which governs the abuse of dominant position. While U.S. trustbusters focused on Microsoft' Windows 95 and Windows 98 desktop operating systems, the European Commission was more concerned with the server market[72].

In order to sustain whether ICANN is violating EC competition principles though, a distinction must be drawn between the place of incorporation and the place where the accused misconduct is occurring. ICANN is first and foremost a US based corporation and initially European laws are not applicable; nevertheless, the fact that the services the corporation provides have a global effect and the role that it is undertaking has international implications, allows the European Commission to take action. The chances are that it would be an easy task - considering that they want to proceed in such an action - for the European Commission to support the allegation that ICANN, via the its role of allocating Internet protocol addresses and parameters, setting the criteria of entering the Root, devising an exclusive dispute resolution Policy and determining the companies that will act as registrars, is in breach of competition rules and particularly of articles 8 and 82 of the EC Treaty.
While the European Commission decides to choose Eurid to run the .eu domain name, ICANN will have to prove now its role and take the appropriate steps for the inclusion of .eu in the global Internet Domain Name System[73]. Unless .eu appears on the proper ISO list at the right time, ICANN is going to have to do some rule twisting to deal with .eu domain name (that is, unless the US Dept of Commerce acts as if it were the US Dept of State and uses its power to dictate .eu into [or out of] existence.)

The competing root community is not so constrained; it can launch the .eu within minutes and thus cause the very fragmentation of the Internet. Therefore, it is more than vital at this stage for ICANN to comply with the European Commission and try to instigate the procedures for the normal adding of the .eu domain name to the authoritative root[ 74].

2. Conclusion

Unfortunately, the issue of Internet Governance and the administration of the Domain Name System has turned into a political debate with ICANN playing a central part. It is an unfortunate situation simply because one of the aspirations of the privatisation process was to avoid political implications. ICANN does not seem, or is not allowed, to act as a private entity and undertake the role that was initially assigned to the corporation. Now that the Internet has reached its peak and its wonders are more predictable it looks as if everybody wants to have a lion's share to its management. Governments are trying to find ways to involve and be part in the administration process.

ICANN has done many mistakes throughout the short years of its existence. Often directed by the US DoC, albeit in a very indirect way, it has lost its initial purpose and gone into areas, which fell outside its scope. The fact is that the corporation is engaged in policy making, that it has a relationship of a strange nature with the US DoC and that it is violating antitrust laws in both sides of the Atlantic. Nevertheless, in this 'road trip', ICANN is not alone; most of the times it has the silent approval of the United States Government and this enables them to act accordingly.

Yet, the US Government's interference is not that unreasonable. The argument that the US officials use is that the United States have conceived the idea of the Internet, they have developed it and expand it and had the means to support its infrastructure; they further purport that the Internet is at the point it is now because the United States had the means to support it. However unreasonable this may seem it has some true; what seems to be an extreme though is the excessive and cunning intrusion that is taking place after the privatisation process. ICANN was created in such a way as to give the DoC the right to interfere and forbid ICANN to take any action on essential issues, unless the DoC gives its consent.

In a similar vein, ICANN shares its part of responsibility. The power that was granted to this private entity and the role that was assigned to its board had the anticipated results. ICANN lost its purpose and ended up exercising governmental functions. The corporation is engaged in policy-making procedures, although it was created as a technical body, it is violating competition rules by abusing its dominant position within the market and it looks as if most of its actions are directed by trademark holders and large multi-national companies.

It would be really interesting to see what the US DoC will do in a case where ICANN is brought before the courts and is found 'guilty as charged' concerning the abovementioned allegations. Until then we should only be patient and hope that the privatisation of the Internet and the Internet itself will succeed.

Notes and References

1 Ellen Rony and Peter Rony, The Domain Name Handbook - High Stakes and Strategies in Cyberspace, R&D Books.

2 See ICANN Fact Sheet, at <> (last modified Mar. 25, 2000). In addition to giving ICANN effective control over the DNS, DoC gave ICANN the authority to make critical decisions relating to Internet Protocol (IP) numbers. In the long run, control over IP numbers is likely to be more important than control over the DNS. The Internet as we know it could function without domain names. It could function, albeit differently, with radically different systems for allocating domain names. It cannot function without a system for the unique allocation of IP numbers. Furthermore, the demand for IP numbers is likely to increase exponentially as designers begin to put IP numbers on everything found in the office or home-even the foods found in a 'smart fridge.' See A. Michael Froomkin, The Death of Privacy, 52 STAN. L. REV. 1461, 1493-94 (2000). ICANN has yet to take any substantial decisions relating to the allocation of IP numbers.

3 Jonathan Zittrain, ICANN: Between the Public and the Private: Comments Before Congress, 14 BERKELEY TECH. L.J. 1071, 1073 (1999).

4 Michael Froomkin, Wrong Turn in Cyberspace: Using ICANN to Route Around the APA and the Constitution, 50 Duke L. J. 17 (2000).

5 Management of Internet Names and Addresses, 63 Fed. Reg. 31,741 (1998) [hereinafter White Paper] also available in <>.

6 See id. At 31,744

7 Supra 4

8 Michael Geist, Public's role in the Net governance threatened, <>.

9 A 'registrar' is a firm that contracts with clients ('registrants') to collect their information and payment in order to make a definitive and unique entry into a database containing all domain names registered in a top-level domain (TLD). A 'registry' maintains this database. Top-level domains are sometimes grouped into 'generic TLDs' (gTLDs), most of which are currently three- or four-letter transnational domains, and 'country code TLDs' (ccTLDs) which are currently two-letter TLDs, most of which exist to serve a national population.

10 See Uniform Domain-Name Dispute Resolution Policy, at <> (Oct. 24, 1999) (defining, for example, the scope of applicable disputes, as well as evidentiary and response requirements)[hereinafter UDRP].

11 Supra 4.

12 See also March 2000 ICANN Meeting in Cairo: ICANN Budget for 2000-2001 Fiscal Year, § II, at <> (Mar. 6, 2000) (reporting an estimate of NSI's payments) [hereinafter 2000-01 Budget].

13 Supra note 4.

14 For more information on the matter see Karl Auerbach v Internet Corporation For Assigned Names and Numbers, Case No. BS 074771, found at <>, accessed on the 29th of May, 2003.

15 DoC calls ICANN's job 'coordination'. White Paper, supra note 5.

16 Letter from ICANN to Rep. Tom Bliley, Chairman, United States House Committee on Commerce, <>.

17 Id.

18 Id.

19 See supra note 4.

20 Karl Auerbach, comments and notes to Konstantinos Komaitis via email, 15th of July, 2002.

21 Supra note 4.

22 Supra note 5.

23 See, The Management of Internet Names and Addresses: Intellectual Property Issues-Final Report of the WIPO Internet Domain Name Process, 23, World Intellectual Property Organization, at <>.

24 See Froomkin M., Comments on ICANN Uniform Dispute Policy: Catalog of Critical Process Failures; Progress on Substance; More Work Needed at <>.

25 Id.

26 Supra note 4.

27 Id.

28 Id.

29 Id.

30 Konstantinos Komaitis, Susan Schiavetta, ICANN's Role in Controlling Information on the Internet, 18th BILETA Conference, April 2003, available at <>.

31 Id.

32 For information concerning the adoption of the new TLDs, <>.

33 Supra note 4.

34 Mission: Technical and/or Policy?, found at <>.

35 Commission of the European Union, Communication from the Commission to the Council and the European Parliament: The Organisation and Management of the Internet International and European Policy Issues 1998-2000, Information Society Promotion Office, <>.

36 Louis Touton, Outline of the ICANN/NSI/Department of Commerce Agreements, Presentation to European Commission Staff, slide 13, European Internet Forum, <>.

37 Supra note 20, 20th of July, 2002.

38 Supra note 4.

39 See NSF9224- Network Information Services Manager(s) for NSFNET and NREN, National Science Foundation, at <>.

40 Department of Commerce: Relationship with the Internet Corporation for Assigned Names and Numbers, United States General Accounting Office, found at <>, accessed on 22nd of May, 2003.

41 31 U.S.C. § 9102.

42 Supra note 39.

43 See Amendment 11 on p. 18.

44 Supra note 5.

45 See ICANN/DOC Memorandum of Understanding, Amendment 1, at <>.

46 Supra note 20, 21st of July, 2002.

47 Millard, Elizabeth., US to Seek Greater Control over the ICANN, <>, (Accessed 09/01/02).

48 See Parket v Brown, 317 U.S. 341 (1943).

49 ICANN and Antitrust, Froomkin, Lemley, at <>.

50 Ibid. For more information see e.g. Beverly v. Network Solutions, Inc., 1998 WL 320829; Watts v. Network Solutions, Inc., 1999 WL 778589; Thomas v. Network Solutions, Inc., 2 F. Supp. 2d 22 (D.D.C. 1998); PGMedia v. Network Solutions, Inc., 51 F. Supp. 2d 389 (S.D.N.Y. 1999).

51 Supra note 44.

52 Testimony, Internet Management, Limited Progress on Privatization Project Makes Outcome Uncertain, <>.

53 15 U.S.C. § 2

54 Supra note 48.

55 See <>.

56 See <>.

57 See <>.

58 While groups such as have a substantial part of gTLDs, they are not used by many people.

59 Supra note 48.

60 See Communiqué of the CCTLD Support Organisation (Information), Meeting in Marina Del Rey, November 12, 2001, at § 5, <>; see also ICANN Threatens the Stability of the Internet,

61 Devin Murphy, ICANN strikes back, refuses to be strong-armed by, Network Briefing Daily, Issue, July 12, 2001, <>.

62 See <>.

63 See <>.

64 Supra note 48.

65 ICANN: The Debate over Internet Governance, 2001, Duke Law and Technology Review, 0002

66 Supra note 48..

67 Id.

68 Management of Internet Names and Addresses, 63 Fed. Reg. 31,741 (1998), found at <>.

69 Supra note 48.

70 'Product Market': market in products, which are interchangeable in the consumers' eyes. 'Geographical Market': i.e. the common market as a whole or a substantial part of it. The substantial part is interpreted accordingly to the goods we are dealing with.

71 Wilcox, Joe, Judge rules Microsoft violated antitrust laws, found at <>, accessed at 28th of May, 2003.

72 Wilcox, Joe, Microsoft faces tougher battle in Europe, found at <>, accessed at 28th of May, 2003.

73 For more information please see <>, visited 2nd of June, 2003.

74 Found at <>, visited 2nd of June, 2003.


1. Books

Halberstam, S. and Brook, J. and Turner, J, (2002), Domain Names, A Practical Guide, (Reed Elsevier UK)

Lloyd, I., (2000), Information Technology Law, (Butterworths).

Grewlich, K., (1999), Governance in 'Cyberspace', Access and Public Interest in Global Communications, (Kluwer Law International).

2. Journals

ICANN: The Debate over Internet Governance, 2001, Duke Law and Technology Review, 0002

3. Conference Proceedings

Komaitis, K. and Schiavetta, S., (2003), ICANN's Role in Controlling Information on the Internet, 18th BILETA Conference, London, UK.

4. Links to Uniform Source Locators (URLs)

































ELJ Home Page JILT Home Page JILT Issue 2002 (3) Search JILT
JILT logo and link to home page