£50 pound notesThe case of Gulati v MGN Ltd [2015] EWHC 1482 (Ch), to which the attention of this blog’s readers has already been drawn here, is a complex but important case. In particular, the fact that no publication to the world at large had occurred in the case of one claimant, and that the element of distress was considerably downplayed because the hacking had occurred unbeknownst to the claimants, forced Mann J to engage with foundational issues in the still-fledging law of privacy in England and Wales. This post will focus on the topic of damages (the sort of losses that are being compensated for) and do so from a mainly private law theory perspective.

Perhaps more clearly than any case before it, whether in privacy or tort more generally, Gulati exposes the frontal clash between two models of understanding the relationship between tort and harm (or wrong and loss) – which, while not limited to breach of privacy, have found in this cause of action a fertile ground to compete on. The important thing, it is argued, is not to mix and match them (something that Mann J. only partially succeeding in doing). The underpinning issues being highly theoretical, we need to start the argument on a high level of generality.

The ‘bipolar’ model

In English law – though not only in English law – the arch-dominant model on which tort law is built is to contrast wrong and loss. The wrong (here the breach of privacy) is the violation of a right (the right to privacy in that case, whatever the exact content of the right might be). Losses, on the other hand, are detriments which flow from the breach. Being consequences of the wrong they are by construction separate from it. These losses, in turn, are either pecuniary or non-pecuniary.

Pecuniary losses mean that the claimant is out of pocket: the law has never had much difficulty with them, provided they can be linked to a right-violation. Non-pecuniary losses are more difficult (in part because they are more ethereal, and in part because they are defined negatively and there exists no universally agreed-upon list of such losses); but essentially they are factual detriments which are not immediately valuable in money. The claimant is worse off in a concrete way but the injury is not to his pocket: rather than his ‘having’ it is his ‘being’ that has been injured in a concrete way.

While the point is not entirely straightforward, it should not be too difficult to accept that these non-pecuniary losses all boil down, in the final analysis, to distress (in a broad and non-technical sense): as Prof. Burrows aptly puts it, ‘ultimately all non-pecuniary loss is concerned with the claimant’s distress or loss of happiness’ (A. Burrows, Remedies for Torts and Breach of Contract (3rd ed, OUP 2004), p. 31). Importantly this concept of distress goes far beyond the narrow label of damages for mental distress used by courts and legal scholars, typically as a residual category, to name one category of non-pecuniary losses among others: rather the argument is that all such non-pecuniary losses (listed in McGregor on Damages (19th ed, Sweet & Maxwell 2014), §5-003 as: (i) ‘physical inconvenience and discomfort’, (ii) ‘pain and suffering and loss of amenities’, (iii) ‘mental distress’, (iv) ‘social discredit’) boil down to a disruption of the claimant’s emotional tranquillity, whatever the labels actually used.

Two applications

Physical injury is a good example to understand how this model works. The claimant’s bodily integrity has been violated: he will (typically) suffer some pecuniary loss (medical expenses, loss of earnings etc) and also some non-pecuniary loss (‘PSLA’ in the lingo: Pain and Suffering and Loss of Amenities). On that model his physical injury is transparent: it is a peg on which losses hang. It needs to be there because without it the loss would not be wrongful in the first place, but the claimant does not get compensated for it; rather he gets compensated for the deleterious consequences.

Applied to privacy, this model regards the privacy-breach as the peg and distress and economic loss as the losses hanged on it. There are, however, at least two important complications. One is that, in breach of privacy cases, there is typically no economic loss pleaded, which puts a lot of stress on what is typically regarded as the junior partner in the pecuniary/non-pecuniary dichotomy, and forces the law to confront questions it is ordinarily adept at sweeping under the rug. The second complication is that, across tort, distress is rarely actual distress; it is generally presumed or deemed distress: how bad the claimant is supposed to feel rather than how he really feels. So deemed in fact that, as is well known, unconscious claimants can recover for loss of amenities (even though it is by nature a form of distress). This complication will be ignored here so as not to make things overly complicated.

To summarise, the dominant model operates in two stages: first the wrong then the losses, which in turn are fundamentally of two and only two sorts: the claimant is out of pocket (pecuniary loss) or he is distressed (non-pecuniary loss).

The ‘unipolar’ model

The other model of tort and harm, which clearly underpins Mann J.’s judgment in Gulati, is to conflate the wrong and the injury. On that view, the loss suffered by the claimant in a breach of privacy case is the loss of privacy itself (a view encapsulated at §111: ‘a right has been infringed, and loss of a kind recognised by the court as wrongful has been caused’). But that is not another type of loss flowing from the breach as per the above model. Indeed loss of privacy does not flow from the breach of privacy in any meaningful sense: it is the breach itself looked at from a different angle. It is not a separate detriment that may or may not follow the wrong: it is necessarily there by virtue of the breach having occurred.

Language inspired by Hugo Grotius (De Iure Belli ac Pacis (1625), §2.17.2) is useful to describe that model: the loss (damnum) is the ‘diminution’ of the claimant’s right, which itself is no different from the infringement of the right hence from the wrong. The loss the claimant has suffered is that he has been wronged; conversely the wrong was the unjustified causation of the relevant loss. On that model the two elements, wrong and loss, are the flip side one of the other.


The immediate question raised by the identification of two models would seem to be, which is better? But that question will have to wait for another day. Suffice it to note, for the record, that this note writer has considerable sympathy for Mann J’s argument (contra the uber-dominant model) that this abstract notion of loss is the better one. In particular it allows us to understand why, like here, it is not only possible but desirable to grant substantive [i.e. not nominal], compensatory [i.e. not vindicatory] damages to claimants whose privacy has been violated, regardless of whether or not they are out of pocket or aggrieved because of the wrong they have suffered (§115: ‘The absence of distress does not that mean that there was any the less an invasion of privacy’; §132: ‘Damages awarded to reflect the infringement are not vindicatory in the sense of Lumba. They are truly compensatory’). This is because privacy is a valuable good in itself, the very reason why the law protects it in the first place.

Rather the point pressed for in this note is one that should be easier to agree upon, namely, that whatever their merits or demerits these two models are grounded in incompatible logics. They are therefore alternatives that cannot and should not be combined. On a principled level, combining them would amount to adding oranges and apples, right-violations and the deleterious consequences thereof, which is impermissible. Practically it would mean counting the same injury twice, which is equally impermissible for obvious reasons.

Let us unpack this argument. If the law compensates the ‘right-diminution’ it should not also compensate for the economic loss that flows from it (by ‘loss’ here is meant the direct or immediate loss, consequential loss being a different story altogether). Robert Stevens’ example of substitutive damages for a smashed car (R. Stevens, Torts and Rights (OUP 2007), p. 61) applies here, even though Prof. Stevens was trying to make a rather different point: if it costs £2,000 to repair the car, the claimant cannot get £2,000 for the diminution of his property right (the value of the infringement being determined with reference to the cost of repair) and then another £2,000 for the economic loss he has suffered: that is evidently the same injury looked at from two different angles.

This is easy to see because values are fairly objective when the loss is pecuniary. It is more difficult to prove with distress but the principle at play is exactly the same. Leaving aside the red herring of aggravated damages (which are now widely, and rightly, regarded as compensatory for a separate injury), the distress and the right-invasion are the same loss.

Again, physical injury is a better starting point because it has this concrete character which breach of privacy lacks. Thus, we know from the judicial Guidelines that ‘total loss of taste and smell’ should lead to compensatory damages ‘in the region of £28,000’ (Judicial College, Guidelines for the assessment of general damages in personal injury cases (12th ed, OUP 2013), p. 8): but is that compensation for distress/PSLA or is it compensation for the physical injury suffered by the claimant (i.e. the infringement of his bodily integrity)? The language of courts and scholars constantly moves between the two. The only possible answer is that it is both at the same time, looked at from different perspectives. The important point, which should not be controversial, is that there is one injury at play, not two.

Application to breach of privacy

To go back to Gulati and breach of privacy, what we see in the judgment is the court caught between a rock and a hard place. It wants to give substantial compensatory damages for breach of privacy in a situation where there is no or little distress. This forces it to switch to what was described as the ‘unipolar’ model of loss: the loss suffered becomes the wrong itself, i.e. the diminution of the claimant’s privacy (itself sometimes described as a type of loss of ‘dignity’ or ‘autonomy’: these are simply different words to describe the same reality at a higher level of generality). In so doing it clearly rejects Matthew Nicklin QC’s argument, which was an almost picture-perfect defence of the bipolar model: all that can be compensated is distress flowing from the invasion, there being no financial loss alleged in the instant case.

The difficulty is that, not having identified the above tension between two conflicting models of wrong and loss, the court only partly repudiates the dominant ‘bipolar’ model, according to which the wrong is transparent and what matters is its consequences, despite its being incompatible with the favoured approach. The lingering presence of this model, rightly identified by Mr Nicklin as underpinning a number of earlier cases (where courts did not have to choose because distress was the gist of the action) can be seen in the fact that, throughout the judgment, the court oscillates between putting the two losses on the same level – i.e. the claimant can get damages for distress and also damages for loss of his privacy – and seeing them as the flipside one of the other.

This leads Mann J. to move constantly between two positions without, it seems, realising that they are mutually exclusive: on the one hand, that there should be damages for distress and for loss of privacy (e.g. §111, end of §130, §143); on the other hand, that it is ‘unnecessary, if not inappropriate’ (§130) to award damages for loss of privacy on top of damages for distress when there is in fact such distress, the suggestion being that this would be double recovery. At §134 Mann J. comes very close to recognising that the distress is the flipside of the privacy-infringement, hence one injury and one compensatory award. In other words the court oscillates between a pure unipolar model (the loss is the infringement of privacy, of which the distress is the reverse side) and a mitch-match of the unipolar and bipolar models (putting loss of privacy and loss of distress on a par).

The present writer’s position would be to say that the loss in all privacy actions is indeed the loss of privacy. Typically this breach of privacy will cause distress but this is irrelevant; it is a typical consequence not an analytical requirement, and the claimant should not get less if he is not distressed (or even incapable of emotions as in the case of juridical persons) and should not get more because he is in fact distressed, or distressed in a more-than-average way – unless and until a separate harm (i.e. the violation of another right) can be identified, which would rightly trigger aggravated damages. But there is no denying that such a view, rooted in a very broad understanding of loss as any violation (‘diminution’) of a right, is a minority position; and adopting it generally would have an enormous ripple effect on the rest of tort law. The important point for now is to identify and accept that there are two irreconcilable logics at play which should not be mixed and matched, even though this is the easy way out and therefore a constant temptation.

Eric Descheemaeker is Reader in European Private Law at the University of Edinburgh. This post builds on a note initially circulated on the Obligations Discussion Group run by Prof. Jason Neyers at the University of Western Ontario.