How do these philosophers help explain...

[Boyd Baumgartner]

How do these philosophers help explain:
<<This gets at the heart of what counts as individuation and how we justify it>>?

The short answer is that philosophy is/has been concerned with foundations; Why we do what we do. I'm sure you'll agree there's a real need for that kind of information in the discipline. If philosophy is concerned with foundations then, it's relevant. When you hear experts in the field call Friction Ridge Examination 'a leap of faith' on national tv, I'd say there's a problem.

What's my evidence of that? Well just a cursory glance at the industry shows that different relevant communities draw on philosophy in their arguments for/against forensic/friction ridge evidence.

1. Frye (1923)
   
   http://en.wikipedia.org/wiki/Frye_standard
   
   Which says:
   

   "evidence could be admitted in court only if "the thing from which the deduction is made" is "sufficiently established to have gained general acceptance in the particular field in which it belongs."
   

   
   
   The origin and domain of reason, to which deduction belongs is philosophy. You can thank Rene Descartes for his contribution of deduction to the field of Philosophy. And it's an indication that the courts are concerned with how we arived at our claims; the foundation of our evidence.
   
   
2. Daubert (1993)
   
   http://scholar.google.com/scholar_case?case=827109112258472814&hl=en&as_sdt=2&as_vis=1&oi=scholarr
   
   Which states in the decision:
   
   

   "Scientific methodology today is based on generating hypotheses and testing them to see if they can be falsified; indeed, this methodology is what distinguishes science from other fields of human inquiry." Green 645. See also C. Hempel, Philosophy of Natural Science 49 (1966) "
   

   
   
   And
   
   

   "("[T]he statements constituting a scientific explanation must be capable of empirical test"); K. Popper, Conjectures and Refutations: The Growth of Scientific Knowledge 37 (5th ed. 1989) ("[T]he criterion of the scientific status of a theory is its falsifiability, or refutability, or testability") (emphasis deleted)."

   
   
   So in these two sentences we see that Carl Hempel, arguably best known for his Raven paradox in dealing with inductive knowledge claims and Karl Popper, who introduced falsification in direct response to Hempel and other members of the Vienna Circle (http://plato.stanford.edu/entries/vienna-circle/) who were propping up confirmation theories as the defacto method of science. Both Hempel and Popper were Philosophers.
   
   
3. Reference Manual on Scientific Evidence, 3rd Edition (NAS)
   
   http://www.nap.edu/catalog/13163.html http://www.au.af.mil/au/awc/awcgate/fjc/manual_sci_evidence.pdf
   
   We can see in the Third Section that again, the appeal to Philosophers and their contributions to what makes science a special enterprise and worthy of consideration in a trial.
   
   

   How Good Science Works (authored by David Goodstein, Ph.D)
   III. Theories of Science, 39
   A. Francis Bacon’s Scientific Method, 39
   B. Karl Popper’s Falsification Theory, 40
   C. Thomas Kuhn’s Paradigm Shifts, 41
   

   
   
4. Sandy L. Zabell, Ph.D., FINGERPRINT EVIDENCE (2005)
   
   http://www.brooklaw.edu/intellectuallife/centerforhealthscienceandpublicpolicy/scienceforjudges/~/media/10AB0B45B8ED4B1886EE41DE9EB3F15B.ashx
   
   
   Which states:
   
   

   Finally, the courts have a role to play as well. Limits should be placed on the testimony of fingerprint examiners (“100 percent positive identification”), so that their testimony reflects the true limits of their expertise. “Whereof one cannot speak, thereof one must remain silent.”
   

   
   
   Which is a quote from Ludwig Wittgenstein's TRACTATUS LOGICO-PHILOSOPHICUS (1921). Wittgenstein happens to be a Philosopher.
   
   
5. Ralph Haber, 90th International Educational Conference of the International Association for Identification, Dallas TX, August 7-13, 2005.
   
   http://www.humanfactorsconsultants.com/presentations.shtml
   
   Presentation: The Science of Individuation and Identification: Commonalities Across Eyewitness, Fingerprint, DNA and Other Identification Sciences
   
   Individuation is also known as a criterion of identity or indiscernibility principle, which states that there can't be separate objects that have all their properties in common. (sound like something we claim when we make an ID?) Although it has its origins in Aristotle, comes back into play in modern times most notably with Arthur Schopenhauer and Gottfried Leibniz, both Philosophers.
   
   
6. Simon Cole, Seventh International Conference on Forensic Inference & Statistics (2008)
   
   http://socialecology.uci.edu/sites/socialecology.uci.edu/files/cv/scole/academ22.pdf
   
   His presentation: “Beyond Individualization, or How Wittgenstein Can Save Forensic
   Identification,” Lausanne, Switzerland, August 21.
   
   Again Wittgenstein who is a Philosopher.
   
   
7. Jay Koehler, Individualization Claims In Forensic Science: Still Unwarranted (2010)
   
   Extremely long link to paper
   
   Which quotes Cole's reference to Wittgenstein in what he calls "The Individualization Fallacy"
   
   
8. NIJ, Fingerprint Sourcebook Chapter 14 (Hey look, it's you Glenn!)
   
   https://ncjrs.gov/pdffiles1/nij/225334.pdf
   
   Which states:
   
   

   Sir Karl Popper (1902–1994) recognized the difficulty of
   defining science. Popper, perhaps one of the most respected and widely known philosophers of science, separated science from nonscience with one simple principle: falsifiability

   
   
   and
   
   

   Exactly what defines a law and exactly what defines a theory is contested within the philosophy of science. In fact, some philosophers of science (Van Fraassen, 1989, pp 180–181) believe that no laws exist at all. However, the majority of modern philosophers of science believe that laws exist and there are two popular competing definitions: systems and universals (Thornton, 2005).

   
   
   You go on to mention Hempel too, but obviously Popper and Hempel are references to the Daubert decision, but I liked that you mention Thornton's mention of Bas Van Frassen. I like Van Frassen's notion of an Inference to the Best Explanation. He's a Philosopher....
   
   
9. Impression & Pattern Evidence Symposium (2012)
   https://www.forensiccoe.org/ipes/agenda.aspx
   
   Title: Recognize, Develop, and Implement: Building on our Foundations

So, we begin by recognizing that Philosophy deals with foundations, I take you through an abbreviated diverse reference of materials citing the importance of philosophy by renowned scholars, scientific organizations, government entities and critics and finally end by noting the most recent professional organizational conference citing the importance of recognizing our foundations. The shortcoming of the list is that the people most often cited do not represent the pinnacle or last word on the topics of interest. The list I gave Tazman about Wittgenstein, Searle, Tarski, Kripke, etc all have expanded the notions.

I'll cut it off here for now making the case that it's relevant and let you determine what to ask next or if you want me to deal more with individuation which I mention briefly in #5.

[John Vanderkolk]

Boyd,

I expected to also see Charles Sanders Peirce in your list of philosophers with his writings on abductive inference.

Has Bernard Lonergan's writings been part of your studies yet? I like his emphasis on experiences, understandings, and judgments being the foundation of human knowing and believing. That is why I brought Jeremy Blackwood to the IAI seminar in Milwaukee last year, to speak on Lonergan.

John Vanderkolk

[Boyd Baumgartner]

John, you know I love Peirce and Pragmatics! I reference abduction in my Philosophy of Friction Ridge Examination video. I also think the pragmatic notions of iconicity and indexicality have huge impacts on feature selection. I always thought it'd be interesting to see your work with Dr. Busey applied to features that have icon, indexical and abstract values to see if the person actually tracks them as such.

With regards to Lonergan, I haven't read any of him, but he looks to be in the Phenomenological tradition of Heidigger. I'll have a look for sure. If you have any specific writings you'd like me to read, send them my way. I'm always up for a good read.

[John Vanderkolk]

Lonergan's classic books are 'Insight, A Study of Human Understanding' and 'Method in Theology'.

[briano]

To get to the point, I feel that rationalism has had more influence on modern psychology. This is mostly due to the ‘active mind’ that rationalists believe in. The active mind ‘acts on information from the senses and gives it meaning that it otherwise would not have. It seems like the part of the brain that is doing this ‘acting upon the information’ is innate. Humans will make associations between events and object without any conscious effort, therefore ‘giving meaning that it otherwise would not have. But even more than this, there is a structure to the way that humans encode information. In cognitive psychology there are models of how information is encoded. Hierarchies are drawn to show how people associate ideas from broad topics down to small characteristics of objects or other things, and although these are theories, there is support for them.

[Tazman]

Since this discussion originally began on a different thread, there has been a shift from specific ethical concerns in casework to the consideration of the fundamental application of philosophy to science. I understand that the concept that science originated in the broader field of philosophy. And I appreciate that philosophy can help us formulate scientific theories and ideals, as well as answering basic ethical dilemmas of application.

But I care little for the aspects of philosophy as contemplated in theoretical science. Those aspects help us little in our day to day jobs. I am more concerned with the application of philosophy to ethics as contemplated in applied science. The original discussion had to do with ethical decision regarding whether to claim and verify exclusions of all candidates on an AFIS list. From there, I took the discussion to the consideration of processing evidence when faced with overwhelming backlog. Does one apply the "one and done" approach to processing methods, or does one apply complete sequential processing to a single case at the cost of allowing the statute of limitations to expire in other cases while applying a philosophy of "thorough and complete" to the case on the bench?

Or to go back to the concept of verification, Popper says that a distinguishing factor of science is falsifiability. But most departments take this concept as meaning that if an identification is not verified, it is not scientific. I would submit that an identification is just as scientific whether it is verified or not. It is the fact that conclusions are verifiable ("falsifiable") that make our discipline a science, not that a specific identification has been verified. Don't get me wrong on this count -- verification is an important quality assurance measure. But the fact that a particular identification or exclusion has not been verified does not make that identification itself any less scientific. It is still verifiable (or falsifiable). To extend that to an AFIS candidate list, verification of exclusions may be used as a quality control measure, but it does not make the process any more scientific. From a practical perspective, then, one must balance conflicting philosophical ideals ("duty" against "greatest good") in reaching the most ethical (or least unethical) balance of conflicting philosophies to a specific casework application.

Can we get back to these specifics for awhile?

[Boyd Baumgartner]

Tazman, I had always planned on going there with this discussion.

Hypothesis testing has it's roots in Platonic dialogue, the dialectic brought to a more contemporary understanding by Hegel (Hegelian dialectic). Popper writes about it HERE). It takes the form thesis - antithesis -> synthesis. Where antithesis is understood to be a critical process that narrows the flaws of the thesis.

Quite literally Hypothesis means 'Under Proposition' meaning there is a phenomenon that falls under the constraints of a statement. This brings in the notion of phenomenology, a Kantian notion that says we do not have access to the 'thing in itself' (noumenon), but rather we only have access to appearances of and experience with the thing (phenomenon).

So a hypothesis then in our discipline, is a statement which is predicated on different observable and relational phenomena in our domain of inquiry (a friction ridge impression). We call that proposition an Evaluation and it's purpose is to declare a statement of identity. This is relatively uncontroversial on it's face as all objects have an identity such that A = A.

The way we go about this is to inventory distinct entities and ascribe to them properties such that an entity is a function of the properties which has been ascribed to it. By inventory I simply mean "declare the existence of", so this an assertive or declarative statement about a ridge event, ridge curvature, etc (any differential disruption to which properties can be ascribed relative to the surrounding environment. This way I'm not soley limited to the level 1/2/3 definitions)

Properties then serve as arguments (e.g. does terminate) which can take free variables (x,y,z,,etc). They are open arguments and are bound by ascribing the property to a quantifier such as a Ridge disruption (The term I'm using to define any information that is used in a Comparison). So I can say TERMINATES(x) and bind that property to a ridge. The existential nature of this statement (stating that it exists) makes it subject to a truth condition. It's either True or False. This makes the argument open to debate and verifiable.

Understanding feature extraction in this manner then allows us to say that these features can be conjunctive, that is to say we can link them together additively and refer to them sequentially. This is accomplished by the fact that they are products of a function as illustrated in the TERMINATES(x) example. This is how we get to recursivity because the output of such a function can serve as inputs to other functions. This builds complexities such that individual elements combine to form complex groups and complex groups combine to form groups of groups.

The result of this is quite literally a network of information or a web of belief (Quine) comprised of a set of smaller beliefs (hypotheses if you will) http://en.wikipedia.org/wiki/Duhem%E2%80%93Quine_thesis. I like to think of it in the way that AFIS Computers generate the relation of features.

skeleton.jpg (43.61 KiB) Viewed 1471 times

If we were to apply this line of thinking to a practical example such as one of the smiley faces from the smiley files we can think of it as:

smiley.jpg (36.79 KiB) Viewed 1470 times

SMILEY(X)

WHERE:

SMILEY((EYE()^EYE()^((MOUTH())

WHERE:

EYES(IS LATERAL((DOT() ^ (DOT()))

WHERE:

DOT(IS LONG() = IS WIDE())

AND

MOUTH(IS CONCAVE(SHORT RIDGE()))

WHERE:

SHORT RIDGE((OPPOSES(TERMINATES () ^ TERMINATES())^(IS LONG() > IS WIDE ()))

Which is a fancy way of saying that I'm perceiving a smiley face and a smiley face is made up of two eyes and a mouth where the eyes are dots and the mouth is a concave short ridge; where a dot is a ridge of equal length and width and a short ridge is a short ridge characterized by the fact that it is longer than it is wide and it has two terminal ends that are on opposite ends.


In this example we can see how the individual elements are ascribed properties which then combine to create a more complex group. In this way finding resemblance to something or naming in general allows us to bind the properties of the item resembled to the group of features in the impression. This is what I was referring to when I discussed semiotic representation in this thread viewtopic.php?f=2&t=1781&p=14576#p14576 and now you can see the formal proof of such representation. The purpose in general, of such an exercise is to allow us to pick out clusters of information in the network of information.

As I stated above, what we get is a sentence with a meaning (semantics) based on how the information is differentially used (pragmatic) and based upon how it's ordered (syntax). All the while predicate calculus (what we're talking about here) recognizes that there is no direct route to semantics from syntax. (Can you say 'no scientific basis for predetermined number of minutiae to call something an Individualization?)
This applies to Friction Ridge Examination in this way: I would make the assertion that value has semantic and syntactic and pragmatic components. We can show this to be intuitively true by just looking at an impression.

LoopUnderPatternArea.jpg (49.01 KiB) Viewed 1466 times

Because the features are pragmatically sparse (they're non differential, aka similar) and syntactically dense (ordered closely together) it takes more quantity to give a sense about them that's meaningful(semantics). The linguistic equivalent would be if I said "I was very, very, very, very, very, very, very, very, very, very tired". After the second or third "very", additional words of the same kind lose their force in adding to the notion of tired. I would assert that comparisons involving this kind of impression would take more information to render an Individualization than one say from the top side of a whorl. An interesting experiment perhaps...

Bring it back to the topic at hand we can see that from the ascribing of properties to entities in a recursively functioning, pragmatic way way we get a existentially testable proposition. We can call that a Thesis. In a comparison we are using that proposition to evaluate resemblence to a rolled impression. We are saying that Proposition L = Proposition K (which is an identity statement) using a set of information (feel free to call it set theory if you like http://plato.stanford.edu/entries/set-theory/).

Undboutdedly this Thesis is going to make some assertions that are not agreed upon by a second person or may not strike the second person with the same semantic force as the first person. This would be the Antithesis, by which certain arguments are going to be challenged and discarded. This is possible because the truth conditions of predicates can be tested.

After the arguments have been reconciled, there will be a synthesis of the two sets of arguments. We would call this a Verified conclusion. As such through the process of reconciling the two sets of information, the outcome should adhere to the following notions (laws of thought (Leibniz/Schopenhauer):

1. A Subject is equal to the sum of its predicates (A = A, or "No Entity without Identity" - Quine)
2. No predicate can be simultaneously attributed and denied to a subject (Ridge event X cannot both exist and not exist)
3. Of every two contradictorily opposite predicates one must belong to every subject (Evidence for the existence of a ridge event is not also evidence against the existence a ridge event)
4. Truth is the reference of a judgment to something outside it as its sufficient reason or ground (If we're using ridge events to judge identity, we may not appeal soley to the ridge events as sufficiency; aka no point standard) (keep in mind this came in 1891)

So, wrapping it up I'm giving you Analytic Philosophy's contribution of Predicate Calculus into how referents derive sense and work in combination recursively to form more complex referents. This is how we give relevance to data, because data without relevance is well.... irrelevant.

[g.]

Boyd,

Thanks for taking the time to answer the question. You have convinced me of two things:
1. For those that love philosophy, esoteric terms, logic functions, abstract theory, etc then this is akin to mental (and verbal) masturbation. It just doesn’t translate to something practical that I (emphasis on me) can personally use. But that’s ok. If it aids you and other Philosophy oriented folks, then it’s another tool to use. I’d personally not use these terms and concepts on the average juror; I’m still not clear if you (emphasis on you) do or not. So to be clear, I am NOT saying this is NOT useful or it’s nonsense. I’m saying, I don’t find it personally helpful in understanding what I do as an analyst and I especially don’t find it helpful to communicate to someone else. Whereas, I do find a Bayesian framework and likelihood ratios useful. Which leads to #2.
2. When I talk to peers about Bayes/Likelihood ratios, the glazing of the eyes and the immediate sense of “???????” reaction I get from others is probably pretty similar to my reaction to these threads. So if I want to get through to folks and show them why I think stats can be useful, I need to find better ways of communicating it. AND THAT was helpful. Thanks!


And again, I’d love to know if you actually speak this way (or close to this way) with jurors when communicating the basis for an individualization to them…if so, I HAVE to get a copy of those transcripts.


g.

[Neville]

The Police Dept may be hard pushed to employ enough people with degrees in philosophy, mathmatic, and forensic science, and yet be willing to take low pay, work in a scummy Police back office, have an artistic flare and photographic skills but still be happy to work on scenes where the rats love to live, fingerprinting smelly bodies, dealing with hard nosed cops and offenders, not to mention hours in front of a computer.
After spending 5-10 years at uni aquiring the skills needed this would be the last job I would want to take on, no matter how sexy CSI may try to make it sound.

[Boyd Baumgartner]

Glenn, lets fast forward through all the responsive insults I could make about your verbal and mental impotence and your 10 yr PhD being the equivalent of a big truck masking some shortcoming and cut to the chase: Who talks like that? Me, to peers when laying foundation, you know the very thing I said in the first post. That just happens to be a huge difference between us, because when I ask you foundational questions about the validity of your assumptions you can't or won't answer them.

The substantive thing here is that I think you've whored your name out on so many publications that you can't keep track of what the papers say.

For instance, The Human Factors paper: http://www.nist.gov/customcf/get_pdf.cfm?pub_id=910745

Which states,

Link between Feature Selection and Source Attribution

The link between variations in feature selection and the examiner’s ultimate decision is not well
understood. Emerging research suggests that there is a relationship,179 but the effect may be
competing with other major interpretative steps, such as interpreting discrepancy, feature
weighting, and thresholds for reaching identification decisions. A better understanding of the link
between feature selection and the ultimate decision would permit best practices to be formulated
and would allow technology to select the most reliable and useful features on which to base
decisions later in the ACE-V process.

What is clear from the research is that in complex cases, examiners who perceive and compare a
greater number of features are more likely to reach a conclusion about source attribution (rather
than determining the print to be inconclusive). Therefore, in complex cases, there is a strong
need for tools or methods to standardize the selection of features. More generally, standardized
methods for feature selection should result in less variation in the features that are selected and
compared, which in turn should lead to more consistency in the decisions. However, even if
decisions were more consistent, they would not necessarily be more accurate. It is possible, for
example, that the best interpreters of prints would be more accurate using their own methods
instead of following a standard protocol. This too, deserves attention and research.

Just only the kind of attention that you can bring, right? If you'd like to get a hold of some transcripts that are entertaining, get the ones from your recent Illinois Frye hearing in Illinois where the case is made that you treat your own research like Manna from Heaven.

If you believe that giving people jelly beans and making Harry Potter references or that pedaling your autograph is the way to enlightenment, well that's your prerogative. One thing is clear though, where you're leading I don't want to go and I'm not alone. I'll continue to talk about these topics because they're fascinating to me, whether you understand them or not.

[Tazman]

Gentlemen, gentlemen!

Please, let's not forget that we're all on the same team here. This is kind of like an Army soldier and a Air Force pilot and an Navy sailor fighting in a bar while the Al Queda thugs are waiting outside with their Russian assault rifles for the three Americans to leave the bar.

Personally, I am nowhere near the scientist that Glenn is, nor am I anywhere near the philosopher that Boyd is. If truth be told, it is a struggle for me to learn from either of you brainiacs, but I am trying. There is plenty of room in fingerprints for both science and philosophy. The two are not mutually exclusive, nor is the town too small for the two of you to walk the streets together.

Let's respect each others' views and argue our points without the personal attacks.

Okay?

[g.]

Ok.... Ummmm....

My point was simple Boyd and not reflective of you at all.
I personally don't get philosophy. That's it. So
discuss away. Good luck.

Cheers

g.