Wednesday, 27 March 2013

Ten ways in which copyright engages freedom of expression

"If the protection of expression is itself a restriction on freedom of expression, then we have entered a metaphysical Wonderland that stands logic on its head, and undermines core, shared global values about personhood." Recording Industry Association of America, October 2014.

"... some restriction on expression is the inherent and intended effect of every grant of copyright". US Supreme Court, Golan v HolderJanuary 2012. 

Reader Alert: this is a very long (3,750 word) post.  If you would prefer to cut to the chase, here are my ten axes along which copyright and freedom of expression intersect:

Slideshow presentation to ORGCon 2013 based on this post:

The recent Donald Ashby (sub nom Ashby Donald) decision of the European Court of Human Rights has revived interest in the relationship between copyright and freedom of expression. The litigation arose because two of the defendant photographers had put on their US website pictures taken by the third at the Paris fashion shows. Under French law that requires the permission of the couture houses. The highest French Court had confirmed their conviction for criminal copyright infringement.

The photographers lost their appeal to the Human Rights Court. But despite the loss the Court’s decision does finally explode any residual myth that copyright has no impact on freedom of expression.

Article 10 of the European Convention on Human Rights protects freedom of expression as a qualified right. In Ashby the Court explicitly held that the appellants’ activities fell within the exercise of the right of freedom of expression and that the conviction interfered with that. It was therefore necessary to consider whether the interference was justified. However the Court held that in this particular case it could not find that either the photographers’ conviction or the financial penalties were a disproportionate interference. Particular account had to be taken of the Contracting State’s broad margins of appreciation for both commercial speech and the protection of rights of others. 

In the subsequent case of Neije and Sunde the ECtHR found that the appellants' criminal convictions for operating the Pirate Bay site interfered with their right to freedom of expression, since they had put in place the means for others to impart and receive information within the meaning of Article 10 of the Convention.  In this case the Court found that there were weighty reasons for the interference and held the complaint to be manifestly ill-founded and inadmissible.

Ashby provides a good opportunity to review the various ways in which copyright and freedom of expression can interact. Some have asserted that copyright and freedom of expression exist on different planes which do not intersect at all. In 2000 Mance L.J. said in Hyde Park Residence Ltd v Yelland: “Copyright does not lie on the same continuum as, nor is it the antithesis of, freedom of expression.”

18 months later Lord Phillips, giving the judgment of the whole Court of Appeal in Ashdown v Telegraph Group Ltd, said the opposite: “Thus copyright is antithetical to freedom of expression. It prevents all, save the owner of the copyright, from expressing information in the form of the literary work protected by the copyright.”

The US Supreme Court in Golan v Holder has recognised that “some restriction on expression is the inherent and intended effect of every grant of copyright.”

Some have argued that, even accepting that copyright does interfere with freedom of expression, the necessary balance between that and other legitimate interests is struck entirely internally within copyright legislation. Copyright is therefore not susceptible to external human rights scrutiny.

In England that argument was put to rest in Ashdown. If any doubt remained at a European human rights level, Ashby has put paid to it. As a matter of EU law, the CJEU in SABAM v Scarlet has stated that intellectual property rights are neither inviolable nor must be absolutely protected, but must be balanced against other fundamental rights including freedom of expression.

Ashby on its facts was an obvious interference with the photographers’ Article 10 rights, even if the ECtHR was unable to find the interference disproportionate. The defendants had put the photographs of the Paris catwalk shows on their public website. The fashion houses were using copyright to try to limit and control public dissemination of images of the shows. The engagement with the right to receive and impart information is clear.

Copyright can engage freedom of expression more subtly. Depending on the scope of copyright and the nature of the remedies against infringers (or in some cases even against non-infringers), when applied to any particular set of facts there are many ways in which copyright can step over the line and disproportionately interfere with Article 10 rights.

It is difficult, perhaps impossible, to identify clear dividing lines between proportionate and disproportionate interference. This is partly because copyright is itself regarded, at least in European human rights law, as a property right that has to be weighed in a balancing exercise with other rights such as freedom of expression. It is also because the degree of interference may result not just from one aspect of copyright, but from the combination of several; and because the impact depends on the factual circumstances of each individual case.

Different features of copyright can engage freedom of expression on different axes. In this sense copyright can perhaps be likened to a mixing console, on which each aspect of copyright that potentially engages freedom of expression is represented by a slide control. One control set to maximum may itself be disproportionate, either per se or in its application to a particular set of facts. Equally the combined effect of several controls set to less than maximum could also be disproportionate.

Take the example of an “originality” slide control. If the slider were set at a level that restricted the use of high level ideas as well as detailed expression, then the interference with the Article 10 right might be disproportionate in itself; but even if not, the interference would be that much the greater if the remedy for infringement were an injunction preventing publication as opposed to a small amount of monetary compensation that permitted continuing dissemination.

Rather than attempt to identify a clear line between proportionality and disproportionality, or examine how rights should be balanced against each other, this article aims to do no more than describe the mixing console: identify each slider whose movement up or down the scale is capable of engaging Article 10 rights to a varying degree.

The focus of this article is on engagement with freedom of expression and nothing else. Arguments of all kinds can be deployed for and against particular aspects of copyright. No given argument necessarily has anything to do with freedom of expression. If it does not, it won’t feature here.

We should not forget some less prominent creatures of statute such as legislation prohibiting circumvention of copyright protection measures, and EU database right (which is turning out in some respects to be more restrictive than copyright). However for the sake of simplicity this article is limited for the most part to copyright. It is written largely from a UK perspective. Copyright in other countries may differ.

The copyright and freedom of expression mixing console

Article 19 of the Universal Declaration of Human Rights states:

“Everyone has the right to freedom of opinion and expression; this right includes freedom to hold opinions without interference and to seek, receive and impart information and ideas through any media and regardless of frontiers.”
In short, the fundamental right of free speech relates to the free flow of ideas, opinion and information.

Our first group of slide controls concerns the extent of copyright and its interaction with each of these three aspects of freedom of speech.



Slider 1: From Expression to Ideas

It is a truism, if not always an illuminating one, that copyright protects expression and not ideas. Jacob J. (as he then was) cautioned in IBCOS Computers Limited v Barclays Mercantile Highland Finance Limited [1994] FSR 275 that while merely taking a sufficiently general idea does not infringe, to take a detailed “idea” may do so. It is a question of degree.

Wherever and however fuzzily this line may be drawn, one reason why it exists is the recognition that all expression, however original, owes something to what has gone before –and the more general the previous knowledge the more likely it is to do so. To insist that all expression should embody only completely original ideas, shorn of anything that has gone before, would be to require silence. If we were to push the expression/idea copyright slider too far, the babble of voices would gradually diminish to nothing. Well of course it wouldn’t, since such a law would lose all legitimacy long before silence fell. But the serious point is that the idea/expression slide control engages freedom of expression more intensely towards the ideas end of the scale.

The proposition that the idea/expression dichotomy engages freedom of expression was recognised in, for instance, the US Supreme Court decision in Eldred v Ashcroft:

“… copyright law contains built-in First Amendment accommodations. … First, it distinguishes between ideas and expression and makes only the latter eligible for copyright protection. … As we said in Harper & Row, this idea/expression dichotomy strike[s] a definitional balance between the First Amendment and the Copyright Act by permitting free communication of facts while still protecting an author’s expression. … Due to this distinction, every idea, theory, and fact in a copyrighted work becomes instantly available for public exploitation at the moment of publication.”
At the other end of the scale it may be said that if expression is protected too narrowly, then creativity is inhibited through lack of sufficient incentive to multiply its fruits. This is a common enough argument. Copyright was described in the Harper & Row decision of the US Supreme Court as ‘the engine of free expression’. This argument engages a fundamental right of freedom of expression only if one regards the State as having a positive obligation to promote freedom of expression, as opposed to simply refraining from interference with it.



Slider 2: From Expression to Opinion

The right of freedom of expression is most robustly protected by international human rights law when opinion, especially political opinion, is at stake. So European human rights law has developed a hierarchy in which individual political expression is more robustly protected than, say, commercial speech. As reiterated in Ashby, the ECHR allows Contracting States a greater margin of appreciation (i.e. allows them more latitude to interfere) with commercial speech.

The potential for copyright to interfere on the opinion axis is high, even when an element of expression is taken. Hence the existence of various exceptions for activities such as criticism and review and for parody. These define circumstances and conditions in and under which it is legitimate to take some element of expression. However, specific exceptions are not always adequate. In the USA the more flexible fair use defence is recognised as an essential part of copyright’s accommodation with the First Amendment:

“We then described the “traditional contours” of copyright protection, i.e., the “idea/expression dichotomy” and the “fair use” defense. Both are recognized in our jurisprudence as “built-in First Amendment accommodations.” (Golan v Holder)
In Europe, for political expression direct recourse has sometimes successfully been made to external Article 10 considerations. In the Netherlands the Nadia Plesner Darfurnica case, although it concerned Community design right rather than copyright, is an example of freedom of political expression trumping the exercise of an intellectual property right.



Slider 3: From Originality to Facts

Works of authorship – literary, artistic, dramatic and musical works - attract copyright only if they are original.

Originality is closely related to infringement. For instance the infringement test as formulated by the EU Court of Justice in Infopaq is whether a part of the author’s own intellectual creation has been taken.

So the level at which the threshold for originality is set affects not only whether a work is protected at all by copyright, but what kind of use of the work the copyright owner can prevent. The degree of use that can be prevented is also affected by whether the infringement threshold is, for instance, the taking of any part or the taking of a substantial part.

Where the work contains factual information, a low originality threshold can protect not just the author’s form of expression, but the facts themselves. Facts are at the most basic level of information that humans may wish to impart or receive. The extract from Eldred v Ashcroft quoted above makes clear the engagement with freedom of expression.

A low originality threshold may overly restrict the ability to recommunicate facts imparted through the means of a copyright work. The potential for this consequence is why exceptions exist such as fair dealing for the purpose of reporting current events. These may be quite limited. For instance in UK law the exception does not apply to photographs.

Beyond copyright the ability to recommunicate facts is increasingly restricted by database right, which is specifically aimed at protecting otherwise unprotectable collections of data. Database right has few exceptions, an investment rather than an originality threshold and (as currently interpreted by the courts) hair trigger infringement criteria.



Slider 4: From Dissemination to Reference

Copyright has always sought to prevent the dissemination of infringing copies. It has typically stopped short of preventing people from referring to the existence of infringing material, whether in a library index, in a footnote to an article, in a newspaper or whatever.

When copyright, in its enthusiasm to prevent the spread of infringing copies, seeks to prevent people from acknowledging their existence or pointing to where they may be found, it engages the right of freedom of speech.

In the digital and online world it does so most clearly when it seeks to characterise linking and similar acts as infringement, or bears on the creation and use of search engines. For activities such as image or video search the engagement may take on further aspects, for instance if effective referencing depends on the ability to make and store a thumbnail of the image or of a frame of video.



Slider 5 From Zero to Perpetuity

It is difficult to identify any particular duration of copyright that triggers engagement with freedom of expression; although if it is accepted that any copyright protection necessarily affects freedom of expression, then the longer the duration the greater the interference. The longest durations of copyright tend to give rise to significant orphan works issues - an especially visible engagement with the freedom to receive information.

Attempts in the USA to challenge copyright extensions on First Amendment grounds have not fared well. In Eldred v Ashcroft term extension to life plus 70 years involving resurrection of expired copyright for some works was held not to offend against the First Amendment. In Golan v Holder introduction of copyright for foreign works previously in the public domain was held not to violate the First Amendment.

Our second group of slide controls concerns who infringes and by doing what.



Slider 6 From Suppliers to Users

The wider the net of infringement is thrown, the greater the scope for copyright to engage with freedom of expression.

Traditionally copyright tended to concentrate on plagiarists or creators who ‘overborrowed’ from the work of others and on manufacturers and distributors.

Copyright bit hard on those at the head of the duplication supply chain: the owners and operators of printing presses, vinyl pressing plants, CD manufactories. It bit less hard on the middlemen between them and the final purchaser. In general they infringed only if they knew they were dealing with infringing copies.

At the point of retail sale it was not an infringement to purchase an infringing item. Even less was it an infringement to read or view an infringing item. Pre-digital, the burden of determining what was and was not infringing was thought not appropriate to place on the user of copyright works.

Digital technology has changed all that. The ubiquity of temporary and transient copies within user devices has provided the opportunity to argue that mere enjoyment of copyright works can infringe, and on a strict liability basis at that.

Enjoyment as infringement is in broad terms the issue currently before the UK Supreme Court in Meltwater (sub nom Public Relations Consultants Association Ltd v The Newspaper Licensing Agency Limited and others). It is a telling indication of how far things have moved in the digital era that the judge at first instance in that case was able to say that the temporary copies exception “cannot have been intended to legitimise all copies made in the course of browsing or users would be permitted to watch pirated films and listen to pirated music.” (emphasis added)

Users and consumers are the least well equipped to determine whether something is protected by copyright, whether the purpose for which it is to be used requires authorisation, and if so whether it is authorised or not. In human rights terms, placing such a liability burden on the user may chill the right of freedom of expression. since the user may be deterred from legitimately accessing information and knowledge.



Slider 7 From Participation to Facilitation

This issue concerns the degree of participation in an infringing activity necessary to be regarded as an infringer. Imposing liability on those who volitionally do the act or cause it to be done is (questions of strict versus fault-based liability apart) uncontroversial. But if liability is extended to those who merely facilitate or enable an infringement, that is likely to have an impact on legitimate activities which otherwise promote the dissemination of lawful material. The right of freedom of expression is consequently engaged.

Similar issues arise where accessory liability rules (such as those of joint tortfeasorship) impose liability on one person for the infringing act of someone else. The English courts have tended to draw a reasonably clear line between facilitation or assistance (non-liable) and involvement of a kind that makes the other person’s infringement one’s own (liable).

A particular area of potential conflict concerns online intermediaries. Conduits, hosts and other similar entities are the engines that drive the free flow of information on the internet. As platforms and pipes they can be regarded as facilitators. Many seek to characterise them as profiting from infringement and duty bound to prevent the use of their services by infringers. Regardless of the rights and wrongs of that debate, it has long been recognised that placing liability burdens on intermediaries has the potential to interfere with freedom of expression. This was specifically recognised in Recitals (9) and (46) of the Electronic Commerce Directive, which provided liability shields for conduits, caches and hosts.

Potential, as much as actual, risk of liability is capable of engaging with freedom of expression if a liability regime incentivises the removal of material that may not in fact infringe. This is often raised as an issue with ‘notice and takedown’ regimes.



Slider 8 From Copying to Communicating

It is natural to think that copyright is about copying. Indeed it is, but not only so. A whole range of acts restricted by copyright now exists. Some types of restricted act are more prone directly to engage freedom of expression than others. One of the most obvious is communication to the public, especially when deployed against linking.

Sir Tim Berners Lee said in the early days of the Web:

“The first amendment to the Constitution of the United States … addresses the right to speak. The right to make reference to something is inherent in that right. On the web, to make reference without making a link is possible but ineffective - like speaking but with a paper bag over your head.”.
The question of linking as infringement is a live issue. It is the subject of the current Svensson reference to the CJEU, in which some rightsowners are asserting a right to prevent others from linking without permission to authorised copyright material on their own sites.

The right of freedom of expression is also engaged in the context of links to unauthorised material. Cases such as Newzbin in the UK have held that at least in certain behavioural contexts some kinds of linking can amount to infringement by unauthorised communication to the public of the linked-to material.

The third group of sliders concerns Remedies

The degree of interference with freedom of expression is not just a question of liability, but also of remedies. The appellants in Ashby complained separately about the conviction for copyright infringement and the amount of the financial penalties.



Slider 9 From Pennies to Prison

The nature of any remedy enforcing an aspect of copyright that engages freedom of speech will bear on the proportionality of the interference with freedom of speech. Here we simply list some relevant factors in assessing the proportionality of remedies against infringers:

- Criminal versus civil liability.

- For criminal liability: imprisonment versus fines.

- Compensatory awards versus aggravated, exemplary or punitive damages.

- Compensatory awards versus fixed (statutory) damages.

- Injunctions against future dissemination and the penalties for breach of an injunction.



Slider 10 From Targeted to Scattergun

Remedies that are not limited to the dissemination of infringing material, but have the potential to interfere with access to legitimate non-infringing material or activities, engage the right of freedom of expression in a particularly direct manner. Suspension or termination of internet access is one obvious example.

The engagement is more apparent when remedies are made available against parties such as intermediaries who do not infringe or, if the law does provide that they infringe, for whom freedom of expression is acutely engaged as a result of the nature of their activities (see Slider 7).

The CJEU in SABAM v Scarlet held that broad indefinite filtering injunctions against internet access providers (and against hosts in SABAM v Netlog) were incompatible with the freedom of expression rights of users of the services:

“that injunction could potentially undermine freedom of information since that system might not distinguish adequately between unlawful content and lawful content, with the result that its introduction could lead to the blocking of lawful communications.



Consequently, it must be held that, in adopting the injunction requiring the ISP to install the contested filtering system, the national court concerned would not be respecting the requirement that a fair balance be struck between the right to intellectual property, on the one hand, and the freedom to conduct business, the right to protection of personal data and the freedom to receive or impart information, on the other."
The UK courts in cases such as Newzbin2, Dramatico v BSkyB and EMI v BSkyB have recognised that blocking injunctions against ISPs engage the freedom of expression rights of internet users.

Where the site requested to be blocked is unrepresented before the court and the defendant ISPs do not contest the injunction, procedural safeguards may be relevant to ensure that the court is fully equipped to evaluate the implications for freedom of expression of internet users.

Sunday, 24 March 2013

1971 revisited – who will be the first Charter martyr?

Should the Crime and Courts Bill become law it looks likely that some publications, perhaps led by the more independent spirits, will decide not to be regulated under the Royal Charter recognition scheme.  The Spectator has said it won’t sign. Ian Hislop has said that Private Eye is unlikely to. 

Not to join the voluntary Charter club of officially respectable news publications would be a brave decision, since it risks a heavy penalty: exemplary damages if found liable for a Charter-related claim (defamation, privacy and some others); or an order to pay an unsuccessful litigant’s legal costs even if the publisher wins. 
Will a non-signatory then refuse to pay?  At this point it gets quite interesting.  Human rights arguments may possibly take the sting out of sanctions for a Charter refusenik.  But if it comes to it, when a publication’s assets are seized - not for breaking hacking laws, or for bribing public officials, or for doing anything else unlawful, but in the final analysis simply because it refused to join the official club - the State coercion that lurks in the shadows behind this scheme steps blinking into the light. 
The notion that this is a voluntary arrangement, that there is some fundamental difference between State regulation and statutory underpinning, is then exposed for the fiction that it is.  We are back in the territory of village Hampdens, Milton, John Wilkes and the rest.  Some courageous little journal (or perhaps, if the clauses remain in their current form, some unfortunate smalltime blogmeister) will have the opportunity to become a Charter martyr.
The last time something comparable happened was not in the 17th century, but 40 years ago.  A plan with remarkable similarities to the Charter, devised to tempt an unruly and irresponsible bunch of institutions into joining a State approved scheme, failed spectacularly when said bunch of irresponsibles refused to play the game that the government of the day had designed for them. 
The scheme was the 1971 Industrial Relations Act.  The irresponsibles were the trade unions.  At the heart of the Act was a voluntary registration scheme.  The purpose of registration was, according to Employment Secretary John (later Lord) Carr: “to make trade unions responsible organisations and with freedom upon that foundation of responsibility to conduct their activities in the ordinary way as responsible representative bodies.” (Hansard, 26 November 1970).   Familiar stuff.
A registered union had to have its rule book approved by the Registrar. Unions had a choice whether or not to register.  But if a union didn’t register it was at risk if its members went on strike.  Unlike a registered union it had no immunity from liability for inducing a breach of contract.  It was thus exposed to injunctions preventing strike action.   
Behind all this were teeth, in the shape of the specially created National Industrial Relations Court, to which the Act gave jurisdiction over labour disputes. 
The corporatist-minded Heath government, whose brainchild the 1971 Act was, never seriously considered that the unions would not register.  But that is what happened.  Many unions, determined to maintain their independence from the State, refused to accept State regulation of their activities and deregistered either from the start or later on during the lengthy saga of resistance to the Act.  They took the loss of immunities on the chin. 
Things got worse for the government.  Strikes, NIRC injunctions, disobedience, refusal to pay fines for contempt of court, strikers queuing up for martyrdom, imprisoned strikers rescued by the Official Solicitor.  The government’s pet industrial relations scheme finally lay in shreds. 
This time around hold-outs will come as less of a surprise, and will not be backed by an organised mass labour movement.  Seizure of assets to pay exemplary damages or costs orders, rather than imprisonment or fines for contempt, are the more likely end-point of the Charter scheme. 
Once the scheme is in place the government will have no way of determining who is on the receiving end of sanctions.  Politically that is a massive hostage to fortune.  Like the 1971 Act and its imprisoned shop stewards, will the Charter eventually founder under the weight of public opinion when the might of the state is deployed not against a multinational publishing empire, but some small magazine or refusenik blog?

Saturday, 23 March 2013

The War on Gossip renewed

Should the Crime and Courts Bill become law in its Levesonised form, it will have the dubious distinction of being the first UK statute in which the word ‘gossip’ appears.  As such, it can perhaps be seen as a new phase in the War on Gossip commenced by Warren and Brandeis in their seminal article “The Right to Privacy”, published in the Harvard Law Review in December 1890.

Most of the article consists of a measured and closely reasoned articulation of a right of privacy from the perspective of the individual.  But it opens with a tirade against the press which contains a more than passing swipe at modern civilisation in general.  While focusing to a degree on the victim, this also displays a positively Reithian mission to elevate the morals of the uneducated classes and save them from the evils of gossip. 
The passage speaks for itself (paragraph breaks inserted):
“Gossip is no longer the resource of the idle and of the vicious, but has become a trade, which is pursued with industry as well as effrontery. To satisfy a prurient taste the details of sexual relations are spread broadcast in the columns of the daily papers. To occupy the indolent, column upon column is filled with idle gossip, which can only be procured by intrusion upon the domestic circle.
The intensity and complexity of life, attendant upon advancing civilization, have rendered necessary some retreat from the world, and man, under the refining influence of culture, has become more sensitive to publicity, so that solitude and privacy have become more essential to the individual; but modern enterprise and invention have, through invasions upon his privacy, subjected him to mental pain and distress, far greater than could be inflicted by mere bodily injury.
Nor is the harm wrought by such invasions confined to the suffering of those who may be the subjects of journalistic or other enterprise. In this, as in other branches of commerce, the supply creates the demand. Each crop of unseemly gossip, thus harvested, becomes the seed of more, and, in direct proportion to its circulation, results in the lowering of social standards and of morality.
Even gossip apparently harmless, when widely and persistently circulated, is potent for evil. It both belittles and perverts. It belittles by inverting the relative importance of things, thus dwarfing the thoughts and aspirations of a people. When personal gossip attains the dignity of print, and crowds the space available for matters of real interest to the community, what wonder that the ignorant and thoughtless mistake its relative importance.
Easy of comprehension, appealing to that weak side of human nature which is never wholly cast down by the misfortunes and frailties of our neighbors, no one can be surprised that it usurps the place of interest in brains capable of other things. Triviality destroys at once robustness of thought and delicacy of feeling. No enthusiasm can flourish, no generous impulse can survive under its blighting influence.”
So the great unwashed have to be rescued from their addiction to gossip.  Whether or not one regards that as paternalistic tripe, who could have predicted that the UK Crime and Courts Bill 2013 would be the vehicle that finally gives statutory force to the War on Gossip?

Friday, 22 March 2013

A handy visual of the Leveson clauses

Summary of the main Leveson clauses in the Crime and Courts Bill.  Some detail omitted (particularly the schedules excluding specific categories of publisher).  E&OE!
 
Click image to view full size chart.





Tuesday, 19 March 2013

Tamiz v Google - the Court of Appeal verdict

The Court of Appeal has delivered its Tamiz v Google judgment on Google’s potential defamation liability for comments posted to a blog hosted on its Blogger platform. 

Although the court ultimately decided that the claim against Google was an abuse of process, the judgment is of most interest for what it said about Google’s position as a publisher at common law and its possible defences as a secondary publisher under the Defamation Act 1996. 
When analysing the decision it is as well to remember that this was a jurisdiction challenge by Google Inc, the US provider of the Blogger platform.  For Google to succeed on the issues under appeal the court would have had to be satisfied that there was no good arguable case against Google – on the face of it a high hurdle. 
At first instance Eady J decided that Google could not be characterised as a publisher at common law, even after being notified of the defamatory statement - so no case could be made against it.  To reverse that decision the Court of Appeal did not have to decide that Google was a publisher, only that there was an arguable case that it was; and then that there was an arguable case that Google was not covered by any available defences.
Eady J’s judgment laid down a bright line precluding defamation liability for at least some types of intermediary.  The position after the Court of Appeal’s decision is fuzzier.   While confirming that pre-notification Google was not a publisher, the Court of Appeal decided that at some point after notification it was arguably a publisher. 
Also arguably, on the facts, Google was not able to avail itself of defences under the 1996 Defamation Act.  The court found it unnecessary to address the possibility of a hosting defence under the ECommerce Directive.
The finding of abuse of process was founded on the short amount of time for which the comments were available in the period after Google could be considered liable as a publisher and before they were removed.  Jurisdiction was declined on that basis. 
So what can we take from the Court of Appeal’s decision?   The three key points are that Google was:
-           Not a publisher pre-notification.
-           Arguably a publisher at some point post-notification
-           Arguably not protected by S1 1996 Act at some point after notification.

Not a publisher pre-notification
Strictly this aspect of the Court of Appeal’s judgment was obiter, since Tamiz’s claim only covered the period after Google was notified of his complaint.  Nevertheless, like Eady J at first instance, the Court of Appeal considered the position pre-notification.

At common law there are two categories of publishers: primary publishers, who are strictly liable for statements they publish; and secondary publishers who, while they still are regarded as having participated in the publication, have at common law an ‘innocent dissemination’ defence available to them.  Someone who merely facilitates publication is not a publisher at all.
The Court of Appeal found that Google’s provision of the Blogger service plainly facilitated publication of the blog and comments.  However its involvement was not such as to make it a primary publisher.  It was not in a position comparable to that of an author, an editor or the corporate proprietor of a newspaper.  Bloggers were plainly independent of Google and did not act in any sense on its behalf or in its name.
The court was less certain that Google’s role was not that of a secondary publisher, expressing itself only ‘very doubtful’ that it was so.   The court’s reasoning on this point on the face of it conflates what one would think should be three separate issues.

             the degree of involvement necessary to be regarded as a publisher (facilitation versus participation),

             the distinction between primary and secondary publisher, and

             the conditions under which a secondary publisher can escape strict liability.

However the same confusion has bedevilled the innocent dissemination cases since at least Vizetelly v Mudie’s Select Library, a century-old case in which Romer LJ and Vaughan Williams LJ got in a tangle about whether innocent dissemination precluded publication or absolved the disseminator from liability for publication.

In any event the Court of Appeal in Tamiz held that the argument that Google was a secondary publisher could get nowhere for the pre-notification period since it could not be said that Google either knew or ought reasonably to have known of the defamatory comments.
Although the Court of Appeal did not address the point directly, there remains a fundamental difficulty in reconciling a finding that Google was not a publisher (whether primary or secondary) with the broad ground on which Morland J considered in 1999 in Godfrey v Demon Internet that Demon Internet was a publisher of Usenet newsgroups hosted by it: that whenever there was a transmission of a defamatory posting from the storage of the defendant's news server, the defendant was a publisher of that posting. 

As I suggested in my commentary on the first instance decision in Tamiz v Google, this looks like the death throes of that aspect of Godfrey v Demon Internet.
A publisher post-notification
At first instance Eady J held that Google was not a publisher even after notification of the existence of a defamatory comment to a blog that it hosted.  He likened Google to the owner of a wall on which people chose to inscribe graffiti, for which the owner was not responsible. 

The Court of Appeal preferred to think of Google as more like the provider of a notice board: one for which, moreover, Google provided a blogger with design tools for his part of the notice board and facilities for displaying advertisements alongside notices; and, according to the Court of Appeal most importantly, Google provided the notice board on terms of its choice and could readily remove or block access to any notice which did not comply with those terms.

As such Google’s blog platform fell within the scope of the reasoning of Byrne v Deane.  In that case from the 1930s an anonymous defamatory notice was affixed to a wall in a golf club and remained there after it was drawn to the attention of the golf club’s secretary.  The golf club proprietor and the secretary were held to have published the notice after notification, on the basis that by failing to remove it they had made themselves responsible for its continuing presence.

The Court of Appeal held that:
“if Google Inc allows defamatory material to remain on a Blogger blog after it has been notified of the presence of that material, it might be inferred to have associated itself with, or to have made itself responsible for, the continued presence of that material on the blog and thereby to have become a publisher of the material.” 
Google argued that the vast difference in scale between the Blogger set-up and the small club-room in Byrne v Deane made such an inference unrealistic and that nobody would view a comment on a blog as something with which Google had associated itself or for which it had made itself responsible by taking no action to remove it after notification of a complaint.

While the court acknowledged that those were matters for argument, they were not decisive in Google’s favour where the court was concerned only with whether there was an arguable case of publication.
One issue that is left uncertain when applying Byrne v Deane is whether it is relevant to consider if the defendant platform has knowledge of the words complained of and their illegality or potential illegality.  Eady J at first instance in Tamiz thought these might be relevant.  The Court of Appeal said that no inference of publication should be drawn until Google had had a reasonable time in which to act to remove the defamatory comments; and that it was arguable that five weeks was sufficiently long for an adverse inference to be drawn against Google. 
However a reasonable time in which to act to remove the defamatory comments is not the same thing as a reasonable time in which to determine actual or potential illegality.  The judgments in Byrne v Deane itself strongly suggest that the trigger is responsibility not for the defamation, but for the continued publication.  There was no suggestion in Byrne v Deane of any need to show that the defendant knew the statement was defamatory or unlawful, or of any grace period in which to determine unlawfulness. 
At any rate a single poem written on a piece of paper stuck a to a golf club wall in August 1935 does seem a shallow foundation on which to construct an entire edifice of answers for a 21st century internet platform that hosts blogs and associated comments amounting to, according to Eady J. at first instance, half a trillion words and 250,000 new words every minute.
Defence under Section 1 Defamation Act 1996
In the event that it was held to have published the statement Google invoked the statutory ‘secondary publisher’ defence under Section 1 Defamation Act 1996.  This defence has three elements: that the defendant:

(a)  Was not the author, editor or publisher of the statement complained of.  Each of these has specifically defined meanings.  In particular a ‘publisher’ is someone whose business is issuing material to the public.  This is completely different from a publisher at common law as discussed above.

(b)  Took reasonable care in relation to its publication.  This provision creates the notorious ‘Catch 22’: how can one take reasonable care without becoming an editor under S 1(a)?

(c)  Did not know, and had no reason to believe, that what he did caused or contributed to the publication of a defamatory statement.

Eady J had held that Google was not a publisher under these provisions.  The Court of Appeal agreed.  It could not sensibly be said to have “issued” the defamatory comments.  Its involvement was of a kind analogous to the example given in S1(3)(e) of someone who should not be regarded as an author, editor or publisher:
“only involved … as the operator of or provider of access to a communications system by means of which the statement is transmitted, or made available, by a person over whom he has no effective control.” 
Eady J considered that by ‘effective control’ it was likely that the draftsman had in mind effective day-to-day control rather than the possibility of intervention in reliance on a contractual term about the permitted content of a web page. 

These comments may be compared with the rulings of Gray J in MORI v BBC (unreported, 17 June 1999 - not cited by either Eady J or the Court of Appeal) concerning the meaning of ‘effective control’ in the live broadcast exception under S1(3)(d).  Although Eady J’s approach was generally similar, they appear to diverge on the possible relevance of a contractual relationship. 
The Court of Appeal agreed with Eady J.  The existence of a contractual term about the content of blogs was not sufficient to give Google effective control over the person who posted the defamatory comments.

Both Eady J and the Court of Appeal concentrated on S1(3)(e) as the closest analogous example to the Blogger host.  However s1(3)(e) is on the face of it aimed more at transmission-type operations.  Section 1(3)(c), which does not contain a “no effective control” limitation, appears to be more obviously aimed at operations in the nature of hosting.   It applies to any person “only involved … in operating or providing any equipment, system or service by means of which the statement is retrieved, copied, distributed or made available in electronic form.” 
While the distinction between S1(3)(c) and S1(3)(e) is by no means clear and they may possibly overlap, the focus on S1(3)(e) for a blogging platform is mystifying.

1996 Act defence pre-notification
For the period prior to notification the Court of Appeal agreed with Eady J that Google would if necessary have a defence under S1.  There was no basis for concluding that Google had not taken reasonable care in relation to the publication (S1(1)(b)), or that it knew or had reason to believe that what he did caused or contributed to the publication of a defamatory statement (S1(1)(c)).  

It is noteworthy that in McGrath v Dawkins, a case not discussed by the Court of Appeal, H.H.J. Moloney Q.C. speculated that it might be reasonable for a wholly automated large website such as Amazon to take no pre-publication steps at all in relation to user-submitted book reviews.  That also seems to be the import of the Court of Appeal’s conclusion in Tamiz as regards the pre-notification period.
1996 Act defence post-notification
For the period after notification the Court of Appeal agreed with Eady J that Google had taken reasonable care in relation to the continued publication of the comments.  Unlike in Godfrey v Demon Internet, where Demon Internet had taken no steps at all after notification, Google had passed on the complaint about the comment to the blogger. 

However the Court of Appeal disagreed with Eady J about S1(1)(c).  For the same reasons as applied to its publication analysis under Byrne v Deane the court held that is was arguable that at some point following notification it knew or had reason to believe that its continued hosting of the material in question caused or contributed to the publication of a defamatory statement.  For this reason the court was not satisfied that Google had an unassailable defence under S1 of the 1996 Act. 
This was the approach of H.H.J. Parkes Q.C. in Davison v Habeeb, which the Court of Appeal preferred to that of Eady J at first instance.  
However Davison v Habeeb concerned statements made in posts by the actual blogger, not in comments to the blogger’s posts.  In Tamiz the blogger was itself, as regards comments, an intermediary who had the ability to remove comments.  It is questionable whether, when the blog platform has notified the complaint to a more proximate intermediary, S1(1)(c) should be applied to the underlying blog platform as strictly as when the complaint is made about postings by the blogger himself. 
In any event the Court of Appeal found only that the point on S1(1)(c) was arguable against Google – not that Google would be unable to rely on the defence.
Conclusion
As regards common law publication and Section 1 of the 1996 Act, the upshot of the Court of Appeal’s decision is that for a blog platform to pass a defamation complaint back to the blogger is not necessarily sufficient to avoid liability.  It may still be at risk of becoming a publisher at common law and of being unable to rely on the 1996 Act. 

This leaves the Electronic Commerce Directive, for damages, as potentially a stronger line of defence for a hosting platform.  It has always been thought to be so, because to preclude reliance on the defence requires actual or constructive knowledge not just of the defamatory nature of the statement, but of illegality.  That requires some information about the claim and possible defences (see Bunt v Tilley). 
However the Directive does not provide a complete defence to proceedings, since it does not prevent a claim for an injunction.  A platform that is keen to remove itself completely from initiated proceedings will still, therefore, have an incentive to remove allegedly defamatory material even if it is not fixed with knowledge of illegality.