Include the Candidate List?

[Tazman]

If strict labor intensity/pragmatic philosophy is what drives the organization . . .

Obviously at some point, the quality of the service you provide begins to diminish due to lack of thoroughness and I think you run into ethical dilemmas. Where is that point, I for sure couldn't pin point it, but I'm sure you would agree?

I do so love philosophy and ethics. Immanual Kant says we must be guided by Duty. Always live up to your Duties and give your all to their performance. In that spirit, you would indeed do complete exclusions with verifications.

But John Stuart Mill says that action is most ethical which provides the greatest good (okay, he actually said happiness) for the greatest number. In that spirit, you would do each case, one method down and dirty ("one and done" I've heard it refered to), verify idents and out the door.

W. D. Ross urges a balance. Each issue considered alone is accompanied by a prima facie duty. But we do not confront the issues in isolation. Therefore, neither Kant nor Mill give a complete picture. To resolve conflicting prima facie duties requires us to balance our duties and responsibilities.

So, to a pure Kantian, we should exclude and verify, but why stop with the top ten candidates? Or the top 20 or the top 100? There is that categorical imperative to exclude, so we should exclude the whole data base and have every exclusion verified.

To a pure Mill adherent, you may run one pristine latent, ask for five candidates, do a cursory glance, and move on to the next case. "Greatest good" would seem to require no backlog be left unattended and all cases be pushed through ASAP.

Ross says there needs to be balance. So, exclusions and verifications on named suspects, but no reason to exclude and verify those on a candidate list of some arbitrary number.

I guess the two biggest questions I would ask you in response would be:

Are you really using ACE-V if you aren't verifying exclusions?

Is your understanding of scientific methodology strictly positivist in that the aim of science is to positively establish truth, or is it to weed out error with the residual conclusion being the best explanation?

I believe for sure it's the second of the two understandings. If that is the case, you would expect to find more error in organizations using 100% ACE-V with the understanding that erring on the side of transparently weeding out error is best case scenario.

Thoughts?

Of course we want to weed out error, but 100% error free is not a scientifically achievable goal. Wouldn't you agree?

So what is an acceptable error rate? I would submit that in the balance of things, exclusions and verifications on an AFIS candidate list is so ultra conservative that the cost in time outweighs the benefit in extra identifications made.

Of course, if you work in an agency where there is never any backlog, then by all means, do the exclusions and verifications.

But if you work in an agency with a year or more backlog where many cases die when the statute of limitations runs out, then you shift the balance more into the realm of greater good by trying to rush more cases through. In that environment, doing exclusions and verifications on all candidates is time better spent moving on to the next case.

[Boyd Baumgartner]

I do so love philosophy and ethics. Immanual Kant says...

I'd say that I'm most in line Alisdair MacIntyre with regards to what we're talking about, namely that there is no morally absolute rationality. I'd agree with Kant in the sense that there's a purposeful direction (you say duty, I say a piecemeal modern understanding of Aristotelian Teleology), but I'm not a utilitarian to the extent that Mills was, as I'd argue that notion of Justice does not supervene from a straightline 'greatest good' roadmap but rather is emergent from a combination of intention and conditions of satisfaction. (Searle)

Of course we want to weed out error, but 100% error free is not a scientifically achievable goal. Wouldn't you agree?

I agree 100% , that's why I say that truth is a residual effect and not an end.

So what is an acceptable error rate? I would submit that in the balance of things, exclusions and verifications on an AFIS candidate list is so ultra conservative that the cost in time outweighs the benefit in extra identifications made.

The acceptable rate of error in my opinion is the one that's the most contemporaneous and most stable. Allow me to elaborate, while we're obviously not 100% certain or have a 0% error rate, if an error does occur it should be caught as close to the time that it happens as possible. There are the obvious hypotheticals about leaving a killer on the street or sending someone to jail wrongly due to an error that speak to such a need. There's also the reliability or stability of the conclusion. Just as most scientific endeavors are peer reviewed, verification in my opinion gestures in the direction of stability. Obviously there can be failings of multiple people and your argument about the number of candidates to review applies to the number of Verifiers in the case which is why I arrive at the notion that Verifications should be scrutiny and not blind. (If you want to blind test as an assurance method that's good, but it doesn't belong soley in the Verification stage (but that's another thread)) Scrutiny then is the intentional of picking apart the justifications that the Examiner has provided in ACE to see if they meet a consensus of other examiners.

[Tazman]

I think we're talking several different layers and levels of ethical responsibility here.

First, the taxpayers. Do they consistently vote for lower taxes while demanding the best services?
Second, the governing body of elected officials. Do they fund the level of services they expect the lab to provide?
Third, the lab management. Do they recognize when employees are providing equitable services?
Finally, you and me, Boyd. Are we doing the best we can within the parameters imposed by the lab management, the elected officials, and the taxpayers?

I think you and I work in different environments. You have a situation I view as luxurious in which you can pursue excellence in each case. I am conditioned to a different workplace in which I have to strive for a more utilitarian balance -- the greatest good for the largest number of users. If I can hit 90% on 50 cases a month, 95% on 25 cases a month, 98% on 15 cases a month, 99% on ten cases per month, I will shoot for the 90% on 50 cases per month and try to get more cases ready for court, even if it is at the expense of not being "thorough and complete" on each case. Many cases will still die in backlog, but by doing my best with the "one and done" philosophy, I can accept that I am doing my best. The ethical responsibility for the unexamined cases has to pass somewhere further up the chain.

What does Alisdair MacIntyre have to say about that?

[Boyd Baumgartner]

First off, I agree with you on the layers and you're right about our different institutional setups (more on that later).

Secondly, MacIntyre is a virtue ethicist so among the three major branches of normative ethics he defaults to the role of virtues as opposed to rules (deontology) or utilitity (consequentialism). With a virtue being defined as a complex multi-track disposition that influences emotion, perception, attitudes, interests, desires, etc. (Call in phenomenological if you want; my note, not his)

MacIntyre did for ethics what Kuhn did for philosophy of science, namely took a historical, social approach and rejected the individual absolutist methods that those philosophers of the Enlightenment sought to ground. If you're familiar with Wittgenstein's private language argument, think of that applied to ethics, there is no such thing as individual ethics. Ethics are conventional, and bound by complex relations and are really kept in check through the brushing up of practice against the community (virtues are social). That being said, virtues internal to the practice will in some regard be institutionally codified in the traditions of those institutions.

So firstly, we would need to identify the virtues of our discipline. I think on a very broad level we would agree that accurate conclusions, performed in a timely manner, disinterested to external influences would be a good place to start our discussion. From a very abbreviated perspective, when evaluating our two organizations, one of the layers we would need to identify then is the role of impression evidence as an investigative tool which also plays into impression evidence as tool in the administration of justice. Specifically these two items speak largely to the virtue of timeliness and go towards the practical nature of reducing backlog and keeping cases from expiring due to the statute of limitations running out. This speaks specifically to the element to which you referred in your earlier post.

The thing about virtue ethics is that there is overlap with pragmaticism in that there is a cashing out of practical actions from them. Where MacIntyre gets it right is that this practical action is institutionally influenced, which explains why your organization differs from mine, becase we have different charters, customs and traditions which influence our workflow and our work product. Is one better than the other? I think they're incommensurable due to our differential histories even though you'll find overlap in our virtues. So, in a very Quninan way, I'm going bring my theory laden perception to the conversation, and you will bring yours. So, my original angle on bringing up the Verifier in bficken's scenario is that a non verified conclusion is pushing the boundaries of ethics due to my virtue ethics mindset in that they are social, whereas that type of conclusion is not social. I've extended this line of thinking to my current perspective which sees ACE-V as a rationally reconstructed, non linguisitic speech act. Conclusions are institutional facts that mirror brute facts and are intentionally packaged and transmitted via notes/charts/etc. to a Verifier who then needs to decode and receive the proper intention (understanding) of the Examiner.

With virtues being described as they are above, and ACE-V understood as a non-linguistic speech act, I think we have the proper avenues identified from which to approach the philosophy of mind behind a pre-linguistic act and the cognitive elements including bias that underlie such endeavors.

Would you agree?

[Tazman]

I like your approach, Boyd. I have not studied McIntyre, but from your comments I gather he is more Aristotelian, i.e., ethics measured by virtuous conduct. Kant = rules, Mill = consequences, but Aristotle was about virtue.

Where would you recommend I start with McIntyre, most importantly pertaining to ethics in the practice of science?

[Boyd Baumgartner]

His contemporary Bibliography on the topic includes:

After Virtue (1981),Whose Justice? Which Rationality? (1988), Three Rival Versions of Moral Enquiry (1990), and Dependent Rational Animals (1999)

If you want After Virtue for free, someone scanned in the whole book and put it on scribd

http://www.scribd.com/doc/33310420/Alasdair-Macintyre-After-Virtue

With regards to his thoughts on philosophy of science, you've got to read him with understanding of the history of philosophy of science as well. I'd say knowing the timeline of Logical Positivism, to Popper to Kuhn and ultimately Imre Lakatos' notion of a Research Programme (which is really a reconciliation of Popper with Kuhn) and knowing that runs parallel to the analytic philosophy timeline of Gotlob Frege, Bertrand Russell, and Ludwig Wittgenstein which then both intersect at Quine, from which emerges Searle, Kripke, and Putnam and ultimately gets you into identity theory (type/type, token/token) which ultimately puts you at a unique intersection of philosophy of mind/language&science. This gets at the heart of what counts as individuation and how we justify it, which is what we are doing in Friction Ridge Examination.

If you want a good resource of material for free, the lectures here http://www.youtube.com/user/SocioPhilosophy include many full university courses from Berkely, Oxford and Yale.

[g.]

Boyd,

OK, I'll bite. How do these philosophers help explain:
<<This gets at the heart of what counts as individuation and how we justify it>>?

Would your court answer include these philosophersand their treatises? What do they have to justify (which I take to mean "support") individuation (which I took to mean, an "individualization")? SInce I know little about this, as I am guessing is true for many readers, what have I been missing that would help me give a better, more understandable and relatable answer to a jury of lay people? If you had a Daubert/Frye hearing is this what you would focus on? (and avoid over-simplified studies on error rates, variance between examiners, differences in interpretation, etc.) to support the accuracy, reliability, and validity of latent print examinations)?

I can't begin to fathom what this court answer would look like. Can you help? (I might suggest a new thread if it's lengthy).

g.