Richard Hill has published an article in the Oxford University Press International Journal of Law and Information Technology. In it he offers his analysis of the 2012 ITRs and some of the criticisms lodged against them. Given his prior position as a senior staff member at the ITU, Mr. Hill is in an excellent position to offer opinions based on his experience in the process leading up to the WCIT and the WCIT itself.
The process leading up to the WCIT and the WCIT itself saw the introduction of a large number of proposed changes to the ITRs. Most of these proposals were not adopted in the final treaty text but they form part of the record of the development of that text. Mr. Hill fails to mention these proposals and ignores their import in his analysis. I consider this significant given Article 32 of the Vienna Convention on the Law of Treaties (quoted below).
Supplementary means of interpretation
Recourse may be had to supplementary means of interpretation, including the preparatory work of the treaty and the circumstances of its conclusion, in order to confirm the meaning resulting from the application of article 31, or to determine the meaning when the interpretation according to article 31:
(a) leaves the meaning ambiguous or obscure; or (b) leads to a result which is manifestly absurd or unreasonable
Rather, Mr. Hill relies solely on Article 31 as the basis for his arguments to conclude that “it is incorrect to conclude that the WCIT outcomes establish a new international regulatory regime for the Internet and give new powers to the ITU”. He goes on to say that they do not “threaten the current multi-stakeholder model for Internet governance or free speech”. From the narrow perspective of Article 31, these statements might be true. However, when one considers preparatory work and the discussions at the WCIT itself, a rather different analysis emerges along with conclusions inconsistent with the narrower approach.
Mr. Hill first addresses criticisms targeted at text in the Preamble. He states:
“The criticized provision in the Preamble states: ‘These Regulations recognize the right of access of Member States to international telecommunication services.’
“It has been stated that this is an unprecedented new clause which is inconsistent with established principles of international law. It is a dilution of human rights as applied to the individual and creates new rights for governments to avoid international sanctions.”
Mr. Hill believes that this statement does not grant any additional rights. He argues that the 2012 ITRs must be interpreted within the constructs of the ITU’s Constitution which permits Member States to “cut off communications with any other Member State”.
However, given that the 2012 ITRs are subsequent to the ITU Convention, it could be argued that the ITRs have precedence and as a consequence, the Constitutional provision to “cut off communications” no longer applies. This certainly would be the interpretation per Article 31 of the Vienna Convention were it not for the clause in the ITU Convention that states “In the case of inconsistency between a provision of this Constitution and a provision of the Convention or of the Administrative Regulations, the Constitution shall prevail”. (The ITRS are Administrative Regulations as defined in the Constitution.)
From this perspective, one sees that the issue might not be as clear cut as Mr. Hill would have us believe. Rather, we have a set of conflicting instruments that will necessarily require interpretation if a conflict ever arises. How that conflict is resolved is far from clear and will likely entail a look at the preparatory process as well as the WCIT itself in order to determine what was intended by those drafting the treaty text.
During the discussion at the WCIT on the inclusion of the controversial “states rights” text, delegates of some Member States supported its inclusion as a means to balance the rights of sovereign states with the rights of individuals (which are now also expressed in the Preamble). While these comments are seen by many as a demonstration of a fundamental misunderstanding of human rights, they also serve to place the inclusion of this text in context. That context is human rights and an agreement between the 89 signatories that human rights might somehow be balanced with sovereign rights in the area of telecommunication.
An interpretation analysis that focuses solely on Article 31 of the Vienna Convention misses this critical point, while one that includes Article 32 takes into consideration treaty preparatory work and the circumstances of its conclusion. This information could be considered in determining which provisions of the ITU’s governing documents prevail. If the 2012 ITRs are determined to have precedence because of their increased scope (human and states rights) and their later adoption, they could be considered to in fact grant a significant right to Member States that accede to the treaty. Seen from this perspective, one can understand why 55 nations chose to no accede.
If the provision in fact does not grant an additional right because of conflict with the Constitution, one wonders why this provision was added and why the possibility of conflict was not raised by ITU legal staff. Those most familiar with the ITU’s foundational documents would have been aware of the possible conflict and were in a position to raise the issue at the conference. The issue was not raised during the Conference and as a consequence the record demonstrates no opposition based on discussion at the time. Instead, the record shows that states favored an expansion of rights for states as well as the concept of balancing state and human rights.
Authorized Operating Agency
Mr. Hill next addresses use of the term “authorized operating agency” and asserts that “there is no reason to think the new ITRs (which apply to exactly the same entities) could create problems. While it is my hope that no problems arise from the change from Recognized Operating Agency (ROA) to authorized operating agency, I do not share Mr. Hill’s confidence regarding possible problems.
There was significant discussion during the preparatory meetings for the WCIT and at the WCIT itself regarding Recognized Operating Agency and Operating Agency (OA). On the one hand, numerous Member States supported a change to OA and others preferred retaining ROA. At the Conference, “compromise text” was proposed and discussed. The addition of the term “public”, while helpful, is not equivalent to “public conveyance” which some had argued for. Had that term been adopted in the final text, I might agree with Mr. Hill’s assessment but as the text stands, supported by the discussion at the WCIT, I find there is considerable opportunity for problems to arise in this area.
Regional Traffic Exchanges (Internet Exchange Points)
Mr. Hill then discusses regional traffic exchange points and concludes “there is general agreement that implementation of national and regional Internet exchange points should be encouraged” and consequently “it is hard to understand why this provisions should be criticized”. If one believes that the ITU should expand its scope and that the ITRs should address technology-specific issues, I might agree with Mr. Hill’s statement. However, I would prefer that the ITU’s scope remain limited and that the ITRs avoid technology-specific references.
Regarding scope, it is clear both from the preparatory work and the WCIT itself that many Member States prefer an expansion of the ITU’s scope. In particular, proposals to change the definition of telecommunication to include ICTs was (and is) especially concerning to many. Formal inclusion of “information technologies” within the scope of the ITU would be a fundamental change, and could present considerable opportunity for confusion and problems.
If we examine “traffic exchange points”, we find that the discussion leading up to the Conference was about Internet Exchange Points (IXPs) and an “accommodation” was made to avoid technology-specific language. However, there can be no doubt that the language in the ITRs is in fact a reference to IXPs and that by its inclusion in the ITRs, the signatories intend to include the Internet within the purview of the ITU. Some find this objectionable and it is surprising (to them) that others find understanding that objection “difficult”.
From the perspective of a high-level, limited, technology-neutral set of ITRs and a limited, perhaps even more limited, role for the ITU one has no difficulty understanding the objection to the proposed text. From the perspective of an increased role for the ITU, understanding is less forthcoming.
Spam and Security
The last of the treaty provisions to be analyzed are those related to security and spam. Mr. Hill points out that the spam provision is non-binding, that several countries already have spam provisions in place, that states are free to do as they wish, that the ITRs are subordinate to theITU Constitution which recognizes human rights, and that Article 1 (Purpose and Scope) contains an overarching provision that excludes content.
It is true that some nations have spam provisions in place. It is equally true that experts generally regard these efforts as ineffective in mitigating spam whereas private efforts of an international scale have proven quite effective. ITR provisions suggesting that state cooperation will have a positive impact on spam mitigation are at best misguided and at worst will draw attention and effort from other demonstrably successful efforts. This would be most unfortunate.
Mr. Hill also correctly points out that states are free to do as they wish with respect to content. What the new provision in the ITRs does is to provide international recognition of that fact and at least some measure of support since there is a call for cooperation “in that sense”. Those wishing to use the ITRs as a pretext for content filtering will certainly overlook the admonition in Article 1.
Perhaps more importantly, a review of the WCIT record will reveal that during the discussion of this provision, several attempts to introduce the content admonition within the text describing “unsolicited bulk electronic communications” were rejected because it was clear that the only way to determine if a message is spam is to examine its content and stating that this provision did not relate to content was absurd.
A quick look at a definition of spam will reveal the need to view content in order to classify a message as spam. The New Oxford American Dictionary defines spam as follows:
(n) Irrelevant or inappropriate messages sent on the Internet to a large number of recipients
(v) Send the same message indiscriminately to (large numbers of recipients) on the Internet
Determining relevance or appropriateness are matters of opinion and judgement that require analysis and review of message content. The group charged with drafting the spam text recognized this and chose to address the absurdity by kicking the problem “upstairs” where a provision in Article 1 was inserted. However, the absurdity remains.
Mr. Hill claims that Article 1 prevails and that content can not be considered in determining if something is spam. But that leaves us with a completely inoperative treaty provision that surely the ITU legal staff would have recognized and raised to the attention of the Conference. That did not happen and as a consequence we are left to determine what was intended.
An article 32 interpretation yields an analysis that reveals significant dialogue on the subject of spam and the desire of many signatories to address it through state cooperation. A number of non-signatories objected to this text for a variety of reasons including its content-based nature. One possible interpretation of the final text might be that the signatories intended for content to be considered and begrudgingly added the Article 1 admonition in an attempt to induce others to accede. Since that failed, and we are left with an absurdity, the provision in question could now be considered to address spam in all its content-filled glory.
This seems to me both a more accurate reflection of events and likely interpretation given available facts and an analysis of those that signed and those that did not. Those signing wished to address spam at the state level whereas many of those not signing did not.
While an argument can be made that the signatories intended that content not be considered in the application of the spam provision, the facts and a fuller analysis demonstrate something quite different.
Mr. Hill’s arguments on the security provision are interspersed with his analysis of the spam provision and my commentary above applies to security as well. In addition, if we review the record of treaty preparation we will see that security discussions included the phrases “cybersecurity and “information security”.
Some claim the phrases are interchangeable. However, having participated in the discussions leading up to the WCIT I find them quite different with those advocating an “information security” regime far more tolerant of human content-related provisions than those preferring the cyber term. These are fundamental differences and as with the spam provision, we can see that many of those that chose to accede to the 2012 ITRs hold this “broader” security interpretation. Those that did not prefer a narrower focus.
What the treaty means in practice likely will be determined by those that operate under it, not those who do not. To see what a number of those states intend, I suggest reviewing WCIT Document 47-E.
The ITU’s Internet Governance Role and Scope
Mr. Hill concludes his analysis of objections to the ITRs and WCIT outcomes with an analysis of Resolution 3 which “invites Member States to discuss Internet-related technical, development, and public-policy issues within the mandate of the ITU at various ITU forums, and to engage with all their stakeholders in this regard”. He concludes that the provision promotes multi-stakeholder consultations. That is certainly one interpretation.
Another interpretation, one more Article 32-like, reveals that the proponents of this Resolution in fact desire, if not demand, a far greater role for the ITU in the Internet and its Governance. Such a move would not promote multi-stakeholder consultations given the limited role available to many stakeholders at the ITU. Membership is required to participate directly. Governments have a controlling and out-sized role at the institution. Delegates are ill-prepared to engage in Internet-related conversation lacking both technical background and practical experience.
While not treaty text, the message from the signatories is clear. They desire a substantial role for the ITU with respect to the Internet and its Governance.
Mr. Hill has provided us with a thoughtful analysis of the the 2012 ITRs and objections to certain provisions. From his perspective, these objections are unfounded and based on an “incorrect” interpretation of the text. Though I disagree with his analysis, I suspect many will agree with his interpretation (perhaps including the Oxford University Press).
I have attempted to review Mr. Hill’s analysis and suggest alternative interpretations of text and to offer a different perspective on the events leading up to and concluding with the WCIT. I find the larger context not only important, but also essential in interpreting the treaty text represented by the 2012 ITRs.