Irony and God

[Suddenly]

Just an observation.

I'm doing a bit of research into the processes used to decide whether a particular piece of "scientific evidence" is admissible in court. One treatise I've been examining is "Modern Scientific Evidence."

Of particular interest to me is forensic identification. The treatise seems to concur with my opinion that most forensic "science" (fingerprinting, ballistics and so on) does not rest on actual science but on untested (or sometimes untestable) assumptions (no two fingerprints are alike, etc.) The authors describe in detail how courts simply dodge the question or pervert the concept of falsifiability in order to allow these kinds of "science" into evidence. The main dodge was to simply declare that since these processes do not follow scientific principles, they aren't science. Since they aren't science, they therefore do not need to follow scientific principles to be considered valid. Which is of course insane, but it sounds a little more scholarly than "we always have allowed this before, so STFU."

All fine and dandy until I reach the section where they discuss "The Nature of Empirical Questions." One section reads "Does God exist?" "Is the death penalty moral?" ... These are ... questions that are not empirical questions and can have no empirical answers. Science cannot help. The first question acts about the supernatural, while emprical questions are confined to the natural world.

(Sec 4-1.2)

So, it seems, the authors criticize the courts for shifting labels to avoid excluding evidence based on their needs, then the same authors use the same process to sheild God from scientific scrutiny.

Sigh.

[Keneke]

And this is why I highly enjoyed Randi's rant on religion.

[elliotfc]

Could it be that things like fingerprints (which can be studied and examined) are literally tangible, while the death penalty/God are ideas/concepts/supernatural?

Faith is invested in the fingerprint deal, and there is nothing wrong with faith. I disagree in criticizing that as a means of evidence; I think the authors were just being snooty/difficult.

-Elliot

[c4ts]

Faith is invested in the fingerprint deal
Explain.

[hgc]

Originally posted by c4ts

Explain. ie., faith that no 2 fingerprints are alike.

What does "alike" or "the same" mean, anyway, in this context? It's all about their usefulness as a reliably distinguishing characteristic for human identification. This probably rests not strictly on faith, as there is a large amount of evidence that fingerprints are really useful in this way.

[elliotfc]

It is assumed, based on overwhelming evidence, that no two people possess the same fingerprints. There is no reason to doubt that overwhelming evidence. Yet it is impossible to test that hypothesis to total satisfaction, and it remains within the realm of possibility that two persons, could, in fact, have the same fingerprints, even though the odds of that probably could not even be calculated.

All of the above leads a person to be comfortable in saying no two people have the same fingerprints. And to make that statement means you have faith in that statement. What's the big deal? Why is the word faith anathema here? What makes that the proverbial ugly stick of a word?

-Elliot

[Suddenly]

Originally posted by hgc
ie., faith that no 2 fingerprints are alike.

What does "alike" or "the same" mean, anyway, in this context? It's all about their usefulness as a reliably distinguishing characteristic for human identification. This probably rests not strictly on faith, as there is a large amount of evidence that fingerprints are really useful in this way.

What it means is they don't want to be bothered with figuring out actual probabilities or doing studies. For example, they admit that DNA evidence has some chance of replication, but often it is a fantastic number like 1 in several million.

Saying something is unique is a way to slip out of having to do all that ugly math and stuff that they have to do w/r/t DNA evidence without sounding like a technical claim is being made.

[Suddenly]

Originally posted by elliotfc
It is assumed, based on overwhelming evidence, that no two people possess the same fingerprints. There is no reason to doubt that overwhelming evidence. Yet it is impossible to test that hypothesis to total satisfaction, and it remains within the realm of possibility that two persons, could, in fact, have the same fingerprints, even though the odds of that probably could not even be calculated.

An assumption based on overwhelming evidence?

Anyway, the problem is that overwhelming evidence that everyone seems to presuppose just flat doesn't exist. Furthermore, even if we have "faith" that it is true, there still is the problem of whether present methods can distinguish with 100% certainty between these fingerprints. Tests to that effect have been less than promising. There were other issues in the fingerprint cases, but since that point is out on a tangent I didn't get into those. My larger point was to do with cognitive dissonance w/r/t the existance of God as opposed to the existance of evidence to support an identification process.

All of the above leads a person to be comfortable in saying no two people have the same fingerprints. And to make that statement means you have faith in that statement. What's the big deal? Why is the word faith anathema here? What makes that the proverbial ugly stick of a word?

Someone going to prison or the death chamber because their "unique" fingerprints turned up in the wrong place is one problem with basing evidence of guilt on "faith."

Faith is by (at least my) definition belief without evidence. This is a bad thing. What you suggest, making a decision based on evidence, I would describe as a reasoned decision based on all available evidence, not one of faith. Problem is that I don't think there really is all that much evidence to support the claim.

[hgc]

Originally posted by Suddenly


What it means is they don't want to be bothered with figuring out actual probabilities or doing studies. For example, they admit that DNA evidence has some chance of replication, but often it is a fantastic number like 1 in several million.

Saying something is unique is a way to slip out of having to do all that ugly math and stuff that they have to do w/r/t DNA evidence without sounding like a technical claim is being made. I don't know that much about it, so I'll take you word for it (on faith). But one question I have is, do they not do the hard data analysis, or do they not have the data? Oh, and another question: What do they mean when they say DNA evidence has some chance of replication? Is that just being careful not to use the language of absolutes? Is it a matter of DNA examination and analysis techniques that could use improvement? Or do they really think that there might be two people who have ever or will ever live who have truly identical DNA (other than identical twins and clones)?

[Suddenly]

Originally posted by hgc
I don't know that much about it, so I'll take you word for it (on faith). But one question I have is, do they not do the hard data analysis, or do they not have the data? Oh, and another question: What do they mean when they say DNA evidence has some chance of replication? Is that just being careful not to use the language of absolutes? Is it a matter of DNA examination and analysis techniques that could use improvement? Or do they really think that there might be two people who have ever or will ever live who have truly identical DNA (other than identical twins and clones)?

It is complicated. Has to do with changing an historically subjective process into one on solid objective ground.

In things like fingerprinting (including a bunch of other fields, but lets stick to this for comparison) classically an examiner would look at the print from the crime scene, look at a print from record, and just eyeball the two, to see if there was "sufficient agreement" to make a match. This was based on no objective factors whatsoever, and the accepted justification was that this person had a lot of "training" and "experience" to do this and maye passed a lot of proficiancy tests.

The opportunity for mistake (or downright fraud) is obvious here.

Then, it shifted towards objectivity when the examiners coined a phrase called "points of comparison." This was an attempt to quantify the number of "matches" needed for "sufficient agreement." This is still a work in progress (I'm speaking quite generally here, BTW). Now, it would seem obvious that a greater number of points of comparison would relate to a smaller chance of a false match, and honest research could come up with an objective standard. As I mention below, this is the case with DNA.

However, many examiners are trying to maintain the simple fiction of uniqueness so that they can keep a sort of infallable image. Admitting a sliding scale based on comparison points would be like admitting they were talking out their butts for the last 100 years about uniqueness. This conservativism is preventing the advancement of the accuracy of the techniques. It is quite maddening in a way.

To compare, DNA came along when courts were more sophisticated, so in order for it to gain acceptance, it had to show some actual numbers. DNA (big simplification here) has what can be loosely considered "points of comparision," and the more of these points of comparison used, the smaller the chance of finding identical items. Research and sound science has boiled this down into mostly objective criteiria, although there is still sometimes a bit of subjectiveness in reading the raw data.

So, the problem with uniqueness isn't exactly that it is factually false, as much as it is irrelevant. We may say that all people are different, but if we use, say, two points of comparision like height and hair color, we cannot say that we are conclusively identifying a particular person. (i.e. "The murderer was 6'4'' and blonde. You are 6'4'' and blonde. Ergo, you are the murderer.) The issue is not so much the nature of the item, but rather the criteria used to identify such item.

[epepke]

Originally posted by Suddenly
[BAnyway, the problem is that overwhelming evidence that everyone seems to presuppose just flat doesn't exist. Furthermore, even if we have "faith" that it is true, there still is the problem of whether present methods can distinguish with 100% certainty between these fingerprints.[/B]

I agree, and I must say that in my estimation, as someone who was a research scientist for 13 years, you are nicely picking up what science is.

The question is not whether a fingerprint is unique. That's close to certain, given a sufficiently rarefied definition of "unique." The real, empirical, scientific question is how possible or probable it is that the fingerprints from two different people, printed under conditions and on media that are less than ideal, captured by a couple of bozos with graphite and cellophane tape who were hired under rules that forbid the hiring of people who are too intelligent, presented to a judge who is probably thinking about golf and six to twelve people who are almost by definition too stupid to get out of jury duty, be judged the same?

I always get a kick out of when someone says that the probability that this DNA does not indicate the right person is one in four billion or something like that. This is just so obviously pablum down the bib, because if nothing else, there is more than a one-in-one-hundred chance that any given person has an identical twin! What, there's no record of an identical twin? Well, babies do get switched in hospitals, and records do get messed up, still with way more chance than one in four billion. Certainly the probability that an $8/hour secretary misfiled the papers is greater than one in a billion, too.

[daenku32]

People are not convicted on Science, but on reasonable evidence. Fingerprints might not be totally unique, but matching ones increase the probability of involvement. If fingerprints don't match, then that evidence is ABSOLUTELY dismissable. You have an absolute towards innocence, which is good.

I don't think any criminal in any fair trial is convicted with a single piece of evidence.

I have no doubt innocent people get convicted all the time, which is why I personally oppose the death penalty. The system just isn't perfect enough to handly life and death situations.