Can things that look like hard facts and indeed are hard facts be socially constructed? Sure. But many people, indeed quite a few scientists, think not. I remember being quite surprised a decade and a half ago when I realized how many facts were indeed socially constructed. It is more obvious that social facts such as crimes and torts are socially constructed. Certain behavior is deemed to constitute a crime, and that is a construction; we (rather; our lawmakers) do it all the time. But facts about existing types of crime can also be socially constructed, and I find that much more problematic. What particularly concerns me here is the casual interpretation of existing categories of crime or tort on internet behavior.
I remember first realising how true it is that hard facts can be socially constructed by reading John Searle’s brilliant book The Construction of Social Reality. There are lots of facts about money. Money is just as real as my house. But it is purely a social construction. That doesn’t mean that “all truths are relative to the truthsayer”, as people such as the late Richard Rorty notoriously claimed. Many psychologists one encounters like to speak of what is “true for you” or “true for him”, but many of them still intuitively know they can be flattened by the bus they hadn’t seen when they try to cross the road and don’t look first, whatever they might think their truth consists in. Still, there is real meat, not just relative meat, in the debate between “realists” and “relativists”. Robert Nozick wrote a fine book, Invariances, on it in 2002, in which he argues truth is relative and tries to show to what. A decade ago, even I made my own minor contribution to the debate around the Sokal hoax and the Bricmont-Sokal review of some sociological thinking about science. But all that is not really my point here.
Before I get to that point, let me make a relevant plug. I recently read Harry Collins’ lastest book, Gravity’s Ghost, which has a brilliant example of a social construction in science. Collins is notorious among some scientists (and some scientific philosophers, I think) for claiming that much of science is a social construction. No, say the scientists, it is about reality and fact (whatever those are, add the philosophers). Harry has studied the gravitational-wave community, the people trying to detect them, for many years, and far from playing one of those “truth-is-what-you-want-it-to-be” sociologists, he looks to have gone native. And he tells a great story. The book reads like a novel, and is quite short. If you want to know how “big science” is done, it is hard to think of a more entertaining introduction. The book also has one of the finest introductions I have read to the difficulties of statistical reasoning as a guide to reality, with insights I cannot recall having read elsewhere. I probably don’t agree with all of it, but I enjoyed reading it and wasn’t in critical mode as I did so.
One of the major problems, perhaps the major problem after that of money, in gravity-wave science, is knowing when you have seen something. Collins has a wonderful account of how a phenomenon was reified by vote. And the vote wasn’t unanimous! (Buy the book! Read it! End of Plug!)
I talk here about a social construction which worries me. Crimes and torts are being constructed out of Internet activity. I don’t mean here crimes and torts as traditionally conceived, which may be committed in ways enabled by internet technology. The Economist once argued against constructing specific internet crimes out of activities which are already proscribed, but which are enabled in different ways by new technology (in the 1990’s; I am sorry that I no longer have the reference). As a classical liberal in my political thinking, I am very wary of the invention of new crimes specific to the internet, or reinterpreting old categories of proscribed behavior in inventive new ways to cover internet activity.
When working in California in the 1980’s, I felt the consequences of the Morris Internet Worm. Robert Tappan Morris Jr. was widely vilified for having released the worm, and eventually convicted of a crime. I felt at the time that that was desperately all wrong. There is no doubt that it was a watershed moment. Companies and organisations were installing and selling actefacts that I felt were not completely fit for purpose. Specifically, large public companies were selling proprietary versions of the Unix operating system, with the Sendmail program compiled with the debug flag set, thus allowing a back door into root privilege on their customers’ computers. I felt that was simply negligent. That was just one example of people being unwise, arguably negligent, and enabling possible damage to innocent others. I was concerned that such things were becoming a major problem on the internet in those days (I long consider this to have been decisively proved!).
What is one going to do about it? Since such habits constitute force majeur, any action must be political – you must bring it inconveniently to people’s attention, because they will ignore you, as they had been doing for years, if you just say “I don’t think this is a good idea” and leave it at that. But inconvenience too many people, and some of them are likely to try to criminalise your activity; force majeur again. It’s a dilemma; there is a fine line.
We now know well the competing philosophies. One (mine) says that Morris was performing what was, at that time, for that act, a legitimate public service. The public needs to be shown how bad things are, once, to start serious public debate and hopefully rectify the situation. That did not turn out to be the majority view. The majority view added up the collective resources spent on combating the worm (stupidly or intelligently, all of it equal), puts it in dollar terms (or equivalent measure) and says to Morris “that is the damage you have caused“. And that is the view that prevails nowadays. That is the view proposed by the U.S. Department of Defence, which has added up all the resources it claimed were compromised by Gary McKinnon’s hack into and through their WWW sites. The fact that they were apparently using insecure software for a site that, by their own arguments, was essential for US national security is apparently no longer considered a significant and culpable act of negligence. I feel strongly that such unfortunate events as the McKinnon hacking episode and attempted extradition would not be happening had we collectively chosen the other reaction to Morris’s escapade some twenty-odd years ago; holding the owners of frangible systems more responsible for the effects of that frangibility.
Which is why I am exercised by what I take to be the invention of a tort. An article by Lauren Weinstein in today’s Risks 26.78 points to an interesting article by a Business-School computer scientist, Panos Ipeirotis, written a year ago, on a scheme for making money from internet advertising.
Someone set up a bunch of domains with benign names to host invisible ads which were then automatically clicked. The ads were of the pay-per-click variety, and the assessment was only made on the “host” site. However, as far as I understand it, the money went elsewhere.
What puzzles me, indeed slightly disturbs me, is that Ipeirotis and Weinstein describe this as “fraud”, and I presume the WSJ, which picked up on the story, does too. Yes, things that are happening and actions that are being taken are not what they seem, for example, an ad on a 0x0 frame cannot be “seen” by any human, the “click” on it is automatic, and a frame that is loaded is not necessarily the same frame that a cursory scan of the source code might lead you to think it was. But all of these things are commonplace in Internet commerce as it is currently practiced.
I wrote down what I think fraud is. Maybe readers would like to try it for themselves before reading below what I wrote, or what others wrote. I wrote
Fraud is a category of human behavior in which one party is led to believe something which is untrue or misleading by a second party, and is thereby led to engage in a valuable-goods or financial transaction whose nature is not in fact what the first party believes it to be; and where the second party benefits and the first party suffers deprivation.
I then looked it up in James’ Introduction to English Law, which is a standard short reference. It is part of the law of contract. There is something called a “representation”, which is a statement, in this case about goods or material involved in a potential transaction. An untrue representation is called a “misrepresentation”. The tort of “fraud” (or “deceit”) is committed when
a person makes a false representation of fact, knowing it to be false, or without believing it to be true, or recklessly, careless whether it be true or false. The false representation must , further, be made with the intention that it is to be acted upon by the party deceived, and if his claim is to succeed this person must prove that he actually did act upon it to his detriment. (Italics in original omitted)
So I got it more or less right, except that the second party does not need to benefit. A few moments thought shows why.
Now, I am sure that the notion of fraud has been extended to electronic actions, such as those made by automated trading devices, in various ways. But it seems to me that a notion of “misrepresentation” would be key to characterising something as fraud. Advertising is rife with misleading representation, but that misreprentation has to do with perceived content, and there is a lot of law based around that. According to the definition in James’, above, a null representation, such as that shown in a 0x0 frame, would not count as a misrepresentation, for it is not untrue; and it is not untrue for it simply is not!
If something is commonplace, such as pop-under windows and suchlike, or dummy frames such as those with size 0x0, then it is surely not clear in what way there is a misrepresentation (in this technical sense). Saying nothing (or putting an ad in a 0x0 frame) even when you know something, does not constitute fraud: caveat emptor. And whether you are paid for “clicks” surely depends on what the contract says it pays for, and I doubt it says that only clicks are paid for which are made by a bona fide human sitting at a keyboard or tapping on hisher iPhone with intent to view. And in a world in which people pay real money for virtual artefacts on games such as Second Life, it is hard to rely on one’s moral intuition to determine when paying money for something or nothing is “OK” and when it is “not OK”.
If we don’t like what this person is doing, and we collectively decide to proscribe it, then let’s approach our lawmakers and persuade them to do so. Let us not put new whines in old torts.