I’ve been working on my draft article Four Models of Fourth Amendment Protection today, and I need your help finding the right word for one of the models.
First, a little bit of background. Part I of the article argues that there are four basic types of arguments that are used in debates of the scope of a “reasonable expectation of privacy.” The idea is that there are four relatively distinct types of claims that are recognized within Fourth Amendment law as legitimate types of arguments on when an expectation of privacy is reasonable. I’m trying to get the four basic arguments down as completely as I can, and for each model I’m trying to find a good descriptive label for it. (The labels are needed to make it much easier to refer to and distinguish the basic arguments.)
Here’s where I need your help. I’m trying to come up with a good label for the line of Fourth Amendment argument that looks at the likelihood that information will be exposed to others. You see this argument often in Fourth Amendment debates. It runs like this: If the chances are pretty low that information might be exposed to others, given prevailing social practices, you might say that this makes an expectation of privacy in that information “reasonable.” It’s reasonable in the sense that a person of good judgment would expect it to remain private. Conversely, if it’s pretty obvious that the information will be exposed to others, you might say that this makes the expectation of privacy unreasonable. I think of this as a sort of actuarial model: it looks at whether someone with an accurate sense of prevailing social practices would think it very likely or less likely that the item would remain private. In other words, it involves a descriptive assessment of the likelihood that something will be exposed to others or the police. The higher the likelihood, the lower the chances of Fourth Amendment protection. (Or at least so the argument goes; sometimes the Supreme Court accepts this line of thinking, and sometimes it doesn’t.)
The question is, what’s the best label for this line of argument? In my initial draft I called it a norms model, on the theory that it was based on prevailing social practices and social norms. But that’s not exactly accurate: the real issue is the likelihood of privacy. Norms can strongly influence that, but they’re not exactly the same. I could call it the actuarial model, but that seems too technical and a little bit dismissive. The chances-that-it-will-stay-hidden model is, um, well, not workable. Any better ideas?
UDPATE: Other ideas I have thought of include the risk assessment model (too easily confused with the assumption of risk rationale in Fourth Amendment law, which is different) and the odds model (too informal).
ANOTHERUPDATE: RIght now I think the winner is the probabilistic model.
The “dissemination” model?
Is “social/societal expectation” model helpful?
[OK Comments: Perhaps. Although it may be too close to the doctrinal test (see my comment to Michael's suggestion below.)]
“reasonable privacy” model?
[OK Comments: The difficulty is that all four of the models are ways of measuring when an expectation of privacy is constitutionally reasonable. So I don't want the labels to sound too much like the doctrinal test, as it seems too leading (as if that model were more correct than the others.)]
The “likelihood” model? That’s from your phrase, “the likelihood that information will be exposed to others.”
The “foresight” model? There, the longer phrase would be “whether a person with reasonable foresight would expect the information to be exposed widely.”
The “probabalistic” model? That’s a bit like your “actuarial,” but it inaccurately suggests that the decision itself is probabalistic.
The “distribution” model? You look at the distribution of the information, both in the statistical sense and in the sense of how widely it’s been spread. (Twists on this one include the “normal distribution” model for a double pun or the “dissemination” model.)
The “indiscretion” model? That is, has the defendant committed an act of indiscretion by revealing the information to the extent he has, if he really wanted it to stay secret?
Risk respecting? Too much alliteration?
The ‘disclosure’ model? Likelihood that information will be disclosed regardless of intent?
I thought of a few others, but they didn’t pass the basic workability test – either too technical, misleading, or plain dumb.
[OK Comments: The difficulty is that this model isn't necessarily about disclosure; it's about the chances of disclosure. But thanks for the possibility]
I understand why you would avoid “expectations model,” given the history of that word in 4th Amendment jurisprudence. “Probabilistic” or “dissemination” might be better choices. What you seem to have in mind is a widening set of circles for the ordinary dissemination of information. Information that tends to stay within the inner range gets the greatest protection; information that makes it to the periphery gets less. How about “bulls-eye model”? “Quantum model”? You could think of it like the shells electrons orbit in.
BTW you should check out Lior Strahilevitz’s Social Networks Theory of Privacy if you haven’t already.
[OK Comments: Yes, Lior's article is great.]
Here are some I can think of off the top of my head:
“likely disclosure model” or “unlikely disclosure model” (depending on your bent)
“diary model” or “private diary model” or “sock drawer model” (not being a lawyer, those might not be good examples, but an example-based name is likely to be remembered, and you didn’t give any to play off of, so I punted)
In all seriousness, perhaps the “public exposure model” or the “risk of public exposure model”
The availability model. That’s how I always remembered State v. Stevens.
This sounds like an “expectation of privacy” type inquiry…or maybe I totally misunderstood your description.
[OK Comments: It's certainly related to that -- I think that's why it is often used. But I'm trying to avoid a label that is too close to the doctrinal test.]
[Edited by OK]
I think you want a short phrase rather than a word: “likelihood of exposure” or “probability of exposure”
Or, you know, you could always go with, “Ann-Bartow-is-a-super-genius-just-thought-I’d-point-that-out” model, I won’t sue you for right of publicity or anything
[OK Comments: "Likeliihood of exposure" just might work. Thanks for the suggestion, Ann.]
How about the “opacity” model – that is, the question of its privacy rests on an reasonable person’s judgement regarding how easy it is to “see” the information.
A “reasonable person” would, in this case, be “someone with an accurate sense of prevailing social practices.”
A variation on Ann’s suggestion:
the exposure risk model
How about the “prevailing practices” model? If all of the other models you give are more normative, you could use the “empirical” model.
“Expected non-disclosure” model. This gives the reader the viewpoint of the person who expected his/her information would not be disclosed.
Why not just use an acronym, like POE (for, as Ann suggests, “probability of exposure”)? (To keep from writing “dormant Commerce Clause doctrine,” I now abbreviate it “DCCD” in all my articles.) Is there a pattern to the other three models? That might help with the naming. You could also seed your discussion of POE with cheesey E.A. Poe references . . . .
The first thing that comes to mind is the “practical expectation model”. Practical in the sense that real-world, societal norms are considered when determining the expectation of exposure.
I would suggest “empirical risk.”
Exposure model? The liklihood that this information is exposed in some other place?
Maybe the Probability Model, or the Probabalistic Model.
A little clunky, but I think it captures the meaning in Odds Model without the informality.
Perhaps the “security model” or the “insecurity model” would be appropriate. In other words, whether our data or communications are private often depends on how “secure” our communications are. Have you taken a look at the following article in the yale law journal and, in particular, its proposed doctrinal test/analytical framework?:
115 Yale L. J. 1086
[OK Comments: Security is one piece of the puzzle, but not the only piece. Yes, I saw that note, although to be candid I didn't find it helpful (for reasons I shared with the author in an e-mail shortly after publication).]
Tweaking Ann’s suggestion: “exposure-probability model.”
Well, these are just variations on previous suggestions, but what about “exposure expectations model” or “expectations of exposure model”?
The first time I scrolled through the comments I found “likelihood of exposure” to be useful and descriptive. If you really want to stretch, it could be the “negligent exposure” model. However for (I suspect) the same reasons you’d want to avoid the fourth amendment doctrinal terms, you might want to avoid using “negligence”; it’s a term in heavy use and one which already totes around a lot of preconceptions for readers.
[Edited by OK]
The anthropological model. Because what you’re doing is undertaking an anthropological study of the expectation of privacy — making a risk assessment as an observer, but not necessarily a participant, of a culture would. It’s more important to capture in the label how one is looking rather than what one is looking at. If that is too formal, then try the “no-you-didn’t!” model.
How about the “risk of exposure” model? The use of the term “risk” captures the connotation that for proponents of this model, “exposure” of truly private information is a negative result that the Fourth Amendment is intended to protect against. It also implies that prior casual distribution of the information by the party asserting a Fourth Amendment claim undermines the Fourth Amendment claim because the party possessing the information can fairly be said to have assumed the risk of exposure.
“Concealibility model:” The flip side of the likelihood of exposure is, the ease of concealing or protecting that which you do not expect or want to be exposed…
I’d say the “disclosure risk model.” Please note your thanks to me in Footnote 1.
“confidence of privacy” model, or “reasonable confidence of privacy” model, or maybe “reasonable confidence” model.
The “predictive” model.
Since the utility of the label depends on what you are trying to distinguish, you would probably get more helpful comments if you could say something about the other models you will discuss without reproducing the article.
They might have already been noted in the comments, but how about: “disclosure probability” model or “exposure probability” model. Or maybe “disclosure expectation” model.
Call it the “traditional expectations” principle. “Expectation” is doing double duty here, although perhaps only for people who know not only English but a little probability theory. Posner would dig it.
How about the “Fourth Amendment Helps Those Who Help Themselves” model?
How about stepping outside the rigid confines of literal descriptions and get creative.
How about “The Buried Treasure model”.
You put it underground, and, in theory, anyone with a shovel could come around and dig it up, but either because you buried it in an obscure location, or really deep, or in a place that people don’t normally go poking around, you don’t think it likely that anyone will happen upon it. Depending on how good your spot it is, your location may or may not be “reasonable”. Access to the treasure’s location is not necessarily restricted, but it is unlikely to be discovered.
Other outside-the-box suggestions:
The squirrel model (who hides his nuts, which are sometimes found by others, sometimes not, but he may reasonably expect they won’t be.)
The Waldo model or Where’s Waldo model (he’s not hiding, per se, but he is in a location where he expects no one is likely to notice him without effort.)
The needle in a haystack model
The unlocked door model (No, it’s not secure, but I didn’t think anyone was likely to come wandering in.)
The [insert quantum physics term] model (Quantum physics is all about probability. I’m sure there is a good metaphor in there somewhere.)
The statistical model (my lone, boring suggestion – but it’s a lot easier to say than “probabalistic”.)
The golden ticket model
I think I’ll stop now, for the good of all.
How about the ‘Logical Deduction Model’ or a more opaque ‘Inherent Deduction Model”. Because you are deducing an expectation based on a logical or inherent assumption.