[EL] Incidents of fraud ID is designed to stop
Schultz, David A.
dschultz at hamline.edu
Wed Aug 6 14:21:46 PDT 2014
Sigh! Have we learned nothing form what seems to be a decade or longer
debate on voter fraud? The issue should be one about epistemology and
ontology but alas it is mostly about faith and metaphysics. What should be
an empirical and testable debate is hardly that. For the most part those
who believe voter fraud exists infer from the absence of evidence that
fraud exists (but is difficult to find) as opposed to inferring from the
absence of evidence that fraud does not really exist in any meaningful
way. Thus, in many ways their position rests upon faith and untestable
propositions that hardly are empirical at all.
For those who care, Let me quote an extended passage from my recent book
ELECTION LAW AND DEMOCRATIC THEORY regarding how to study the issue. I
strongly recommend all of you to read it and the book.
How can one study voter fraud? To answer this, there is a preliminary
threshold question. What can one make of Posner’s assertion that because it
is difficult to detect voter fraud it is hard to gather real evidence of
its extent? Whatever the current status of knowledge or evidence about
fraud, one thing is certain—this is an empirical question. A hallmark of
modern science and social science is that it is empirical and
evidence-based. It requires the formulation of testable propositions
subject to rigorous methodological tools and the use of data. This is the
core of what the scientific method is.
But even more specifically, (social) scientific research requires
empirical proof to test propositions, and the form of that testing lies in
falsification. As Karl Popper pointed out the hallmark of scientific and
social scientific research is falsification: “But I shall certainly admit a
system as empirical or scientific only if it is capable of being tested by
experience. These considerations suggest that not the verifiability but the
falsifiability of a system is to be taken as a criterion of demarcation.”
Demarcation, for Popper, is separating empirical from non-empirical
theories (such as metaphysics). What Popper is contending, and modern
science has adopted his approach, is that empirical studies are premised
not upon the confirmation of propositions but upon falsifying them.
Scientific inquiry is inductive: “[It p]asses from singular statements …
such as accounts of the results of observations or experiments, to
universal statements, such as hypotheses or theories.” It is generalizing
from specific instances to broader propositions. We generalize from a
survey sample to a broader population; we generalize from a few lab
experiments or clinical trials to broader conclusions about drug efficacy
or reaction. The same is true with social science claims: discrete data are
aggregated and then generalized to produce claims about the world, whether
they be about politics or some other social phenomenon.
Propositions incapable of empirical falsification are not scientific.
Some types of arguments—such as deduction—may be based on logic, while
others are held as belief or faith. This is true for example, with
statements about God. Determining exactly what counts as evidence of God’s
existence is contentious, but, at least since Immanuel Kant in the late
eighteenth century, the general argument has been that it is impossible to
make scientific statements about God’s existence. Instead, it is a matter
of faith.
The reference to God gets at the question of the difference between
“evidence of absence” and “absence of evidence.” This distinction connects
back to voter fraud. Is the evidence of absence an indication that voter
fraud is not a significant problem, or is the absence of evidence simply a
sign, as Posner and others assert, that there is not enough information at
hand due to detection problems and therefore we cannot conclude that fraud
does not exist?
In some cases, evidence of absence is conclusive. A biopsy may reveal
no cancer, indicating a tumor to be benign. An autopsy may demonstrate no
bullet holes, therefore indicating that the victim did not die as a result
of a gunshot wound. In the nineteenth century, the famous Michaelson-Morley
1887 physics experiments with light demonstrated that aether does not
exist. This failed experiment was critical to providing the support Albert
Einstein needed for construction of his special theory of relativity in
1905. Thus in some cases absence is evidence, but absence is still
empirical and the failure to find something constitutes falsification of a
claim. This is evidence of absence.
But in some instances absence of evidence is not conclusive. Failure to
find evidence of God’s existence has been taken not to prove that a divine
being does not exist, but simply for some that we have not yet found
evidence for a supreme being. Those who argue that voter fraud exists take
the latter view—absence of evidence is not proof that it does not exist.
Instead, they offer the absence of evidence as proof of how hard it is to
detect voter fraud. However, what can we really infer from an absence of
evidence—that voter fraud exists in a far more significant level than
thought, or that there really is very little fraud? Minnite asserts that
the problem with Posner and many claims about fraud is that it is an effort
to prove the negative. By that, how do we prove the existence of something
that we cannot detect?
What does all this mean for voter fraud studies? First, the debate has
been cast the wrong way. Minnite and Schultz, as noted above, examine voter
fraud studies and conclude that the absence of evidence is proof that
something does not exist, at least to a large extent. Posner and Fund reach
the opposite conclusion? Who is correct?
To resolve the debate it is important to restate claims about voter
fraud in terms of testable (falsifiable) claims. The arguments need to be
framed in ways that allow it to be tested. One needs to be able to
generalize from specific instances of fraud to be able to make broader
assertions. Are the few instances that are detected a good sample
indicative of a broader population of fraud?
What would be a testable voter fraud claim? As one reads Fund, Smith,
Posner, and others who assert its widespread existence, several claims seem
to emerge. First, we do not know the extent of fraud since we do not have
voter photo ID. This claim recognizes the lack of evidence and asserts the
solution is to institute photo ID. Once in place, it would reveal the
extent of attempted fraud. However, for those who also contend the fraud
exists, once photo ID is instituted, the lack of increase in detected fraud
is proof that the additional measure (photo ID) works to prevent fraud.
While this may be true, proponents cannot have it both ways. The arguments
are a circular logic at best; they both cannot be true at the same time.
Moreover, they may not be empirical and testable propositions capable of
refutation or falsification.
First, think about what would be valid evidence to document fraud. Paul
Wyckoff describes a hierarchy of empirical evidence based on its
reliability. At the top of the evidence chart are controlled experiments
where studies can manipulate critical variables and test them. At the
bottom are case studies and anecdotes where there is almost no ability to
test and verify facts and control variables. A critical issue in evaluating
qualities or types of evidence is the ability to generalize from them to
reach some larger propositions. In other words, do the case studies,
anecdotes, or experiments serve as good samples that represent the larger
population? Generally simple stories or anecdotes are weak evidence because
one cannot easily use them to make inferences about broader trends or
populations unless other evidence or reasons are offered to support drawing
broader conclusions. Additionally, simple stories, unless corroborated, are
like hearsay and are unable to withstand truth tests to ascertain their
credibility and reliability.
On Wed, Aug 6, 2014 at 3:14 PM, Justin Levitt <levittj at lls.edu> wrote:
> I actually share Barnaby's concerns about voter registration problems --
> and particularly the impact that fake forms or sloppy forms or duplicate
> forms have on election officials. (It's one of the reasons I'm very
> sympathetic to calls for a more global change to voter registration that
> would make the process cheaper, more accurate, more expansive, and more
> secure all at the same time.)
>
> But -- just as one example -- I'm not aware of any fake registration form
> submitted by bad-apple ACORN canvassers getting paid for registration
> drives where a ballot was ever cast in the fake name. Ever. (If someone's
> got an example, again, I'd love it.) Which means that what Barnaby's
> talking about is the benefit that ID has in stopping potential fraudulent
> ballots that haven't really materialized in volume yet.
>
> In a real analysis, that would get added to the mix of costs and
> benefits. Stopping lots of actual ongoing fraud is extremely important;
> stopping a trickle of ongoing fraud is important, but comparatively less
> dire; stopping potential fraud is important, but less dire still. That all
> has to get weighed against costs: the harshest ID laws cause real problems
> for some eligible citizens; less harsh laws cause fewer problems; and some
> states have figured out ways to ask people to prove that they are who they
> say they are while causing no problems at all, by accommodating those who
> might not have particular types of documents.
>
> I'm all for policies that prevent potential fraudulent ballots that
> haven't materialized, as long as those policies don't actually cause
> problems of their own. (I take it that this is actually Barnaby's position
> with respect to campaign finance: if we can prevent the potential for
> official corruption without treading on First Amendment rights, that would
> be less worrisome ... but that without scads of what he views to be actual
> corruption, just going after potential problems is unjustified when the
> preventive mechanism has real costs.)
>
> I live in a city where there are millions of houses with millions of glass
> windows. Comparatively few of those houses have iron bars or expensive
> security systems. Those amount to millions of invitations to potential
> burglaries or home invasions. In the neighborhoods where burglaries and
> home invasions are consistent problems, I can understand why residents bar
> their windows -- despite the fact that the bars cost money and can be
> unsightly, the benefits make the costs worth bearing. In the neighborhoods
> where burglaries and home invasions are only potential problems -- even
> though anyone *could* break in at any time -- I can understand why
> residents *don't* bar their windows, but instead take lower-cost measures
> that seem to address the potential problem sufficiently.
>
> Voter registration fraud happens, and when it does, it's a real problem.
> I'm all for well-tailored solutions to that issue (and the other forms of
> fraud Barnaby mentions). But rules requiring ID at the polls don't
> actually stop bad registration forms. And in areas where bad registration
> forms haven't actually led to fraudulent ballots, I don't know that the
> costs of the harshest rules requiring ID at the polls live up to the
> limited benefits of stopping a potential that hasn't materialized.
>
> Justin
>
> --
> Justin Levitt
> Professor of Law
> Loyola Law School | Los Angeles
> 919 Albany St.
> Los Angeles, CA 90015213-736-7417justin.levitt at lls.edussrn.com/author=698321
>
> On 8/6/2014 11:24 AM, BZall at aol.com wrote:
>
> Justin knows that I respect him as a researcher and writer, but that I
> disagree with many of his characterizations and conclusions.
>
> Here, for example, is a brief that I filed in the current Kobach v EAC
> appeal, documenting, through numerous, specific video and news reports,
> "specific and credible" voter registration fraud:
> http://moritzlaw.osu.edu/electionlaw/litigation/documents/Kobach105.pdf.
> (I have repeatedly put up my Crawford brief dealing with voter
> impersonation fraud, and Justin has acknowledged the incidents I
> documented, so I won't do that again.)
>
> The brief illustrates the singular difference between my analyses and
> Justin's: if there is an ambiguity, no matter how small, Justin would
> "bet" that this is the result of error, not fraud. He may be right. But
> defining "fraud" as only an intentional attempt to deceive the system by a
> person participating in an effort to change the outcome of an election --
> which seems to be what Justin tends to do -- downplays the systemic effects
> of such things as Las Vegas unions threatening illegal immigrants with
> deportation unless they vote, and so on. Was it fraud for an organization
> to send pre-populated voter registration documents to dead dogs, as was
> repeatedly done? Was it fraud for an alien to register to vote -- before
> they received citizenship -- even if they received it afterwards? Was it
> fraud for an organization to intentionally register aliens to vote -- the
> Orange County District Attorney reported that 61% of their voter
> registrations were "illegal" -- even if the number who ultimately voted was
> not enough to sway that particular election? Was it fraud to send out
> millions of pre-populated voter registration forms, counting on over-worked
> election officials to weed out the duplicates and the false ones? "It's up
> to [the people who receive] them to obey all ... laws," Voter Participation
> Center's Page Gardner told the Washington Post. Even assuming the purest of
> motives doesn't make up for the fact that the actions put, in the kindest
> of descriptions, substantial stress on a system whose components rely on
> affirmations and signatures (see, e.g. InterTribal Council, upholding the
> EAC's decision to force states to rely solely on a signature). And to say
> that requiring voter ID would not weed out dogs and most illegal immigrants
> is, to put it mildly, odd.
>
> None of these "specific and credible" incidents are in dispute, but
> Justin's characterizations make them seem as though they are negligible in
> effect. This is defining away the problem, rather than analyzing what the
> problem is and what the consequences are. There may have been billions of
> votes cast, and the vast, vast majority may have been legitimate, but under
> Purcell, the Supreme Court notes that the public perception of fraud in the
> system has adverse consequences which are probably much more significant
> than the raw numbers of those reported frauds.
>
> Of the 1.3 million voter registration applications ACORN turned in, only
> 450,000 were legitimate. Reports like that, about which there is no dispute
> over the facts, just the motivation, fuel public distrust. So saying only
> 31 reports of voter impersonation (which may be true, as defined, or even
> in the abstract) among billions, rings hollow. Saying voter ID will not
> help either of those problems is another "thud."
>
> If researchers are wondering about the persistence of the American
> public's concern about fraud, it may stem from repeated attempts to tell
> them: "Nothing happening here. Move along", when something is clearly
> happening. If the explanation looks like minimizing the effect on public
> confidence, the research design is flawed. And so is the Brennan
> advocacy-oriented website, which contains only briefs that confirm their
> view.
>
> Barnaby Zall
> Of Counsel
> Weinberg, Jacobs & Tolani, LLP
> 10411 Motor City Drive, Suite 500
> Bethesda, MD 20817
> 301-231-6943 (direct dial)
> bzall at aol.com
>
> In a message dated 8/6/2014 1:09:03 P.M. Eastern Daylight Time,
> levittj at lls.edu writes:
>
> By the way, in the piece
> <http://www.washingtonpost.com/blogs/wonkblog/wp/2014/08/06/a-comprehensive-investigation-of-voter-impersonation-finds-31-credible-incidents-out-of-one-billion-ballots-cast/>
> about fraud to which Rick linked, I listed the incidents I know of since
> 2000 that reflect credible allegations of the type of fraud ID was designed
> to stop (to Brad's repeated point
> <http://department-lists.uci.edu/pipermail/law-election/2014-July/009759.html>,
> I agree that it's possible that there may be sporadic incidental effects on
> some other mistake or misconduct). And I mentioned that I'd welcome
> additional information.
>
> Specifically, I said that "I am a researcher, and so I am interested in a
> thorough list: if you have credible information about a specific individual
> whose vote was stolen by an impersonator at the polls, please tell me.
> Specific and credible means just that. Not — please — examples like this
> <http://www.washingtonpost.com/blogs/the-fix/wp/2014/06/27/the-manchurian-candidate-of-oklahoma/>.
> And if you have information about an incident below that indicates that it
> was error rather than fraud, please tell me that as well."
>
> That's a sincere request.
>
> --
> Justin Levitt
> Professor of Law
> Loyola Law School | Los Angeles
> 919 Albany St.
> Los Angeles, CA 90015213-736-7417justin.levitt at lls.edussrn.com/author=698321
>
> On 8/6/2014 8:22 AM, Rick Hasen wrote:
>
> “A comprehensive investigation of voter impersonation finds 31
> credible incidents out of one billion ballots cast”
> <http://electionlawblog.org/?p=64062>
> Posted on August 6, 2014 8:15 am <http://electionlawblog.org/?p=64062> by Rick
> Hasen <http://electionlawblog.org/?author=3>
>
> Justin Levitt
> <http://www.washingtonpost.com/blogs/wonkblog/wp/2014/08/06/a-comprehensive-investigation-of-voter-impersonation-finds-31-credible-incidents-out-of-one-billion-ballots-cast/>
> for WonkBlog:
>
> I’ve been tracking allegations of fraud
> <https://web.archive.org/web/20070622014244/http:/truthaboutfraud.org/index.html> for
> years now
> <http://www.brennancenter.org/sites/default/files/analysis/The%20Truth%20About%20Voter%20Fraud.pdf>,
> including the fraud ID laws are designed to stop. In 2008, when the Supreme
> Court weighed in on voter ID, I looked at every single allegation
> <http://www.brennancenter.org/page/-/Democracy/Analysis%20of%20Crawford%20Allegations.pdf>
> put before the Court. And since then, I’ve been following reports wherever
> they crop up.
>
> To be clear, I’m not just talking about prosecutions. I track any
> specific, credible allegation that someone may have pretended to be someone
> else at the polls, in any way that an ID law could fix.
>
> So far, I’ve found about 31 different incidents (some of which involve
> multiple ballots) since 2000, anywhere in the country. If you want to check
> my work, you can read a comprehensive list of the incidents below.
>
> To put this in perspective, the 31 incidents below come in the context of
> general, primary, special, and municipal elections from 2000 through 2014.
> In general and primary elections alone, more than 1 billion ballots
> <http://www.fec.gov/general/library.shtml> were cast in that period.
>
> Some of these 31 incidents have been thoroughly investigated (including
> some prosecutions). But many have not. Based on how other claims have
> turned out, I’d bet that some of the 31 will end up debunked: a problem
> with matching people from one big computer list to another, or a data entry
> error, or confusion between two different people with the same name, or
> someone signing in on the wrong line of a pollbook.
>
> Posted in election administration <http://electionlawblog.org/?cat=18>, fraudulent
> fraud squad <http://electionlawblog.org/?cat=8>, The Voting Wars
> <http://electionlawblog.org/?cat=60>, voter id
> <http://electionlawblog.org/?cat=9>
>
>
>
>
> _______________________________________________
> Law-election mailing list
> Law-election at department-lists.uci.edu
> http://department-lists.uci.edu/mailman/listinfo/law-election
>
>
>
> _______________________________________________
> Law-election mailing list
> Law-election at department-lists.uci.edu
> http://department-lists.uci.edu/mailman/listinfo/law-election
>
--
David Schultz, Professor
Editor, Journal of Public Affairs Education (JPAE)
Hamline University
Department of Political Science
1536 Hewitt Ave
MS B 1805
St. Paul, Minnesota 55104
651.523.2858 (voice)
651.523.3170 (fax)
http://davidschultz.efoliomn.com/
http://works.bepress.com/david_schultz/
http://schultzstake.blogspot.com/
Twitter: @ProfDSchultz
My latest book: Election Law and Democratic Theory, Ashgate Publishing
http://www.ashgate.com/isbn/9780754675433
FacultyRow SuperProfessor, 2012, 2013
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://webshare.law.ucla.edu/Listservs/law-election/attachments/20140806/d06f229c/attachment.html>
View list directory