Lott and Glassman on Felon Voting

Lott and James Glassman have a piece in the New York Post arguing that felons should not be allowed to vote. Well, I can't claim to know anything about the issue (for that, see Kevin Drum and Julie Saltman), but this is John "98%" Lott and James "36,000" Glassman, so you just know their numbers are going to be dodgy. And sure enough, in the first paragraph we find:

A bill to guarantee that millions of convicted murderers, rapists and armed robbers can vote.

That sounded like a lot to me, so I thought I'd check. I found a table giving a breakdown of felony convictions by crime type and it seems that 1% are for murder, 1% are for rape, and 2% are for armed robbery. These sorts of felons are not typical (drug offenses are the most common category), and certainly there are not millions (or even hundreds of thousands) of them.

I was also impressed by the entirely circular argument they offered for not allowing felons to vote:

Why shouldn't felons be able to vote if they have paid their debt to society? Simply because society believes that the debt includes a prohibition on voting.

And how we know that society believes they shouldn't vote? Why obviously because society has laws against them voting. In other words, they shouldn't be allowed to vote because they aren't allowed to vote.


More like this

You've met Mary Rosh and maximcl and Washingtonian. Now meet Bob H and Tom H and Sam and Kevin H and Too bad Tim is not very accurate and Gregg. Yes, Lott created a whole army of sockpuppets that he uses to post comments on my blog. At first he would use just one sock to make a point or dispute…
The Australian published a letter to the editor the day after Lott's piece on laser pointers: John Lott (Opinion, 24/3) claims an Australian academic with a laser pointer would cause panic. I'm an Australian academic and when I use a laser pointer it does not cause a panic. Lott…
Lott has posted some criticism of Chris Mooney's article. Let's see how many errors he has successfully identified: 1) Paraphrasing claim from the Chronicle of Higher Education stating that the "coding errors had not been reviewed by a third party." I was never asked by the…
Richard Lempert comments on why he found Lott's results implausible when they first came out: To give another example, long before other research called their results into question, it was common sense that made me suspicious of John Lott and David Mustard's claim in the Journal of Legal Studies…

similar to the debate about wether felons should be able to have their firearm possession rights restored. It seems to me that it ought to depend on what the felony was, as you seem to ellude to with the voting aspect. Certainly, there is some room for discrimination among felonies.

The whole notion of restricting rights based on criminal history is something I find interesting. If convicted of crimes, do you revoke voting rights? Gun rights? Right to assemble and associate with others? Once you restrict one, it pretty much allows reign to restrict others.

After the outrageous fiddling of the electoral rolls in Florida leading up to the 2000 election, I'm surprised that anyone can advance the argument that it is a good thing to exclude felons.

If we look at the history of this, it is almost always part of the South's tactics for disenfranchising African-Americans and preserving power structures.

If you do deny them citizens rights, perhaps they need no longer pay tax. The US independence was won on a slogan of "No Taxation Without Representation".

By Don Wigan (not verified) on 02 Mar 2005 #permalink

Hate to burst your Southern boogieman bubble (look under you bed Don, it's Jefferson Davis playing 'Skynard on his iPod!), but denying felons the right to vote has its origins in the English Common Law. Commit a felony, and you give up most of your rights and privileges.

At any rate, and at the risk of putting words in Tim's mouth, you're missing the point of his post. Namely that Lott and Glassman have a very dysfunctional relationship with the truth.

Spare us your intellectualizing, Soctt. Your link says nothing about voting rights, but even if it did, how would that refute the rather obvious notion that these rules are a ploy to deny representation to the underprivileged? Or, the fact that felon voting bans are more common in the south?

In Florida, for example, 31% of black men, or 16% of all blacks voters, cannot vote because of felony convictions.

By Pro bono mathe… (not verified) on 02 Mar 2005 #permalink

but denying felons the right to vote has its origins in the English Common Law.
By a twist of history, it was removed from the English lawbook by statute in 1870, shortly after "the South" lost the American Civil War. What's your excuse for the last 135 years?

In Florida, for example, 31% of black men, or 16% of all blacks voters, cannot vote because of felony convictions.

Tough bananas. Don't commit felonies. Being "underprivilaged" matters not. Besides, it's not as if being banned from voting keeps you from being politically active. You can still spend your time and money on political action, which is potentially more effective than having a single vote.

I am especially careful to make sure that I don't inadvertently break some difficult to know law that could result in a felony charge because that would result in loss of my gun possesion rights, which are somewhat more important to me than my voting rights.

Table 7 on Tim's link appears to show 600,000 currently incarcerated violent-crime felons, most of them fitting into the murder/manslaughter/rape/armed robbery category. Released ex-felons should take that figure into the millions. Am I missing something?

FWIW, I despise our criminal "justice" system, and think released felons should be able to vote.

Nine states outside the south, Arizona, Connecticut, Delaware, Iowa, Kentucky, Maryland, Nebraska, Nevada, Wyoming don't automatically restore voting rights to felons. Two of them, Iowa and Nebraska, are just as restrictive as Florida. How again is this all the result of the "South's tactics for disenfranchising African-Americans"?

I don't see why people who have been released from prison should be prevented from voting. On the other hand, it is reasonable that those who are serving time should not vote - as is the case with many prisoners in Australia. It is difficult to see how lawbreakers should vote for lawmakers.

By Steve Edwards (not verified) on 03 Mar 2005 #permalink

Tim, your last argument is a bit sloppy. It misrepresents the degree of circular behaviour; 'society believes' is reflected in law, voted on by elected representatives. The prisoners and ex-prisoners are not elected themselves, then held from exercising a representative vote by hostile opposition. Representatives are elected to act for ALL their constituents, and when laws are passed improving human rights (eg giving voting rights for women or blacks) they do so despite the disenfranchised not having elected the representatives. The new community standard is thus promoted to law despite the fact that those who can't vote are less well represented.

Ben: <<Tough bananas. Don't commit felonies.>>

Ben do you seriously beleive that the proportion of black felons who get caught and convicted is the same as for white felons?
Poor people are more likely to be convicted that rich people simply because they can't afford the same standard of legal
counsel. Blacks, on average, are poorer than whites.
So an apparently non-racial-based law has the effect of excluding a disproportionate number of black people - kind of like those laws many of the same states used to have on the books saying you had to get ten already-registered voters to vouch for your identity before you could register to vote or requiring that one of your grandparents had been on the electoral roll.

By Ian Gould (not verified) on 03 Mar 2005 #permalink

We don't generally remove people's rights unless we think they abuse them, somehow, do we? I'd agree it was sensible to restrict someone's gun ownership rights if they have committed a crime with a gun, but on what pretext can one remove voting rights? Voting for the wrong party? voting for the wrong politics? Having the wrong interests at heart?

Makes little sense to me.

"oor people are more likely to be convicted that rich people simply because they can't afford the same standard of legal counsel."

Which reminds me, how's Noelle Bush doing these days? Reckon she's been struck off Florida's electoral role for a felony bust?

They recycled a variant of this piece on TCS, Tim. In comments I pointed here. I trust the discussion will remain on topic and civil. D

I see 173,000 individuals convicted in 2000 of violent felonies. I don't understand Lambert's so-called math ability. If states on average prevent these people from voting for 10 years on average, that means about 1.7 million violent felons are prevented from voting. Violent aggravated assaults (hardly a trivial offense like the drug violations that Lambert emphasizes) make up most of these. With life prohibitions on voting you could get many millions. You just don't compare it for one year. Lambert needs some real help doing simple math.

Tom H is the one with a reading comprehension problem. Lott and Glassman did not talk about all "violent felonies" but "A bill to guarantee that millions of convicted murderers, rapists and armed robbers can vote". Readers might want to note the blindingly obvious difference and why it makes Tim's point correct. This would also explain Tom H's inability to understand Tim's math ability.

By Mark Schaffer (not verified) on 04 Mar 2005 #permalink

Not all states bar felons from voting, so as well as not being not what Lott and Glassman said, Tom's number is also incorrect. If you want an estimate of the number of violent felons that would be enfranchised, take the percentage of felony convictions that are violent and multiply that by the 4.7 million that would get the vote.

Okay, Table 6 in Tim's link is the critical one. Tim's analysis is accurate looking exclusively at murder, rape, and armed robbery, which is a fair way to look at it because that's what the authors Lott and Grassman said. If you add all the violent crimes (which I think would also be fair), the ratio is 173/924, or 173% according to the guy who critizes the Lancet study. Using old math, I get 19% of felony convictions are violent. Nineteen percent of 4.7 million is 893,000, less than half of the "millions" we were told would destroy us all, but not a completely tiny estimate either.

"A bill to guarantee that millions of convicted murderers, rapists and armed robbers can vote." This does not say that those who currently can't vote will be able to do so. What it says is that these criminals will be guaranteed to vote. Guaranteed by the federal law. If there are about 90,000 criminals per year that committed murder, manslaughter, rape, & robbery, then that implies that there are 1.8 million over 20 years. By the way, the year 2000 in Table 6 is a low year for violent crime. Convictions per year would have been much higher in the 1970s, 1980's, 1990s. Thus if anyone cares to go back and look at the numbers, then it would appear Lott&Glassman are quite correct.

Hi Tim,

Wouldn't you also want to account for repeat offenders over the course of however many years Tom H wants to drag this out? This would also reduce the actual number who would vote. And of that percentage only some fraction would actually go to the polls and of that fraction some would even be stupid enough to vote for the neo-cons! I think there was another discussion on your site regarding someone's propensity to repeatedly punch themselve's in the face. I think we may have a new winner here with Tom H!

By Mark Schaffer (not verified) on 05 Mar 2005 #permalink

Tom H, you're making a different comparison. L&G state they're looking at disenfranchised ex-felons and saying millions of those 4.7 millions are murderers/rapists/armed robbers (or being generous to L&G, are violent felons). You can try and "improve" their argument by applying it to all violent ex-felons from all states, but that doesn't contradict Tim's work in pointing out the errors in L&G's statements.

Personally, I think they were just being sloppy - they probably meant all violent crimes, and were too lazy to figure out if the total would top 2 million. Being off by 60% is nothing to the author of "Dow 36,000".

Brian S., You are mixing two different paragraphs together. One is a reference to what Hillary C. is saying about all people who are currently banned. Another is the number of people who would be "guaranteed" the right to vote no matter what the states try to do in the future, and it involves a subset of the people who commit certain felonies. Face it, Lambert blew it again by only looking at numbers for ONE SINGLE YEAR (a year with a very relatively low crime rate) and ignoring what the word "guarantee" means. Lambert's discussion did even begin to acknowledge that you would want to look at more than one year and that picking a year where the crime rates were half (I assume) what they were in earlier years might be misleading to extrapolate from.

Even using Tom H's strained interpretation of Lott & Glassman's sentence, it still isn't millions. Acccording to report on Felony Sentences in State Courts, 1992 (1992 is when crime was at its peak), there were 12,548 convictions for homicide, 21,855 for rape, and 13,810 for armed robbery. That's 50,000 a year. Over 20 years we get at most one million and that's certain to be an overestimate because crime was lower in other years and we're double counting repeat offenders.
And I don't think any reader is likely to interpret the sentence Tom's way, since the article is about the felons who will be given the vote, not the ones who already have it.


When we allow for multiple convictions, the number of felons would obviously be even lower than the number of felonies.

By Ian Gould (not verified) on 05 Mar 2005 #permalink

I probably missing something trivial, but "guarantee" sounds like it protects all felons ability to vote. Guarantee doesn't only sound like change from not vote to vote. There are plenty of other words and phrases that Lott&Glassman could have used if that is what they meant. Also Tom's 20 years seems like a pretty conservative estimate since the guarantee would apply to felons of all ages. While some felons might only live for a few years after release others may live for a long time. I don't know what the number to use is, though 20 seems low.

I will give you the benefit of the doubt (though it does bias the results in your desired direction), but why are you using "Felon Sentences in State Courts" in your most recent post and not all felons. Are you claiming that those convicted of federal felonies are able to vote in states? Is that true?

My reading of the earlier posts says that if you place felony manslaughter with murder or believe that they are referring to violent felons generally Lott&Glassman have way more numbers than needed to make their claim.

While Bob and Tom H venture into what-the-meaning-of-"is"-is territory (and lying about that too - manslaughter is already part of <1% murders, Bob, and yes, felons do vote in most states), this is all quite irrelevant.

The fundamental point is that Lott and Glassman deliberately leave the reader with the impression that the 4.7 million felons that Clinton talks about are all murderers, rapists and armed robbers. They know it is not true, they know that the reader will think it is so, they leave themselves some wriggle room just in case someone calls them on this trick.

Lies. Just standard right-wing modus operandi.

By Pro bono mathe… (not verified) on 06 Mar 2005 #permalink

"Don't commit felonies. "

If the Jeb Bush-Katherine Harris 2000 electoral roll in Florida is any thing to go by, you should also ensure that you don't have the same (or even a similar-sounding) name.

The argument about exclusion sounds very similar to the old one, now not often used, that them folks in the poorer crime-ridden urban areas, not even owning their own places, should not have the same rights as us decent law-abiding, church-going country folks with an investment in the community.

By Don Wigan (not verified) on 06 Mar 2005 #permalink

Bob, Lott made the claim, it's his job to support it with evidence. Where is the data that you allege shows there are millions of murderers and rapists who would be guarenteed the vote? Why don't you ask Lott and tell us what you find?

In itself, in the abstract, I would have no problem with denying the right to vote to people who have been convicted of serious crimes. The problem is the second-order effect: that if felony convictions remove voting rights permanently, this creates a powerful political incentive to convert more and more crimes into felonies in order to disenfranchise people you don't like. And it seems to me that this is exactly what is happening in states that do this.

Note also that the company responsible for the 2000 Florida felon list, which arguably swung the presidential election, was ChoicePoint, currently embroiled in an enormous identity-theft scandal. Even if you support the principle, giving power over the franchise to these clowns is a terrible idea.

"Pro bono" wrote: "While Bob and Tom H venture into what-the-meaning-of-"is"-is territory (and lying about that too - manslaughter is already part of <1% murders, Bob, and yes, felons do vote in most states), this is all quite irrelevant." The claim is that there is an inaccuracy and the only way that claim can be made is misinterpreting the word "guarantee." I don't read anyone explaining "guarantee" doesn't mean exactly what is consistent with the meaning that make Lott&Glassman's statement correct (that is the law will ensure that these criminals can vote in the future).

The claim that "Lott and Glassman deliberately leave the reader" is strange because it requires that readers not realize that there are two separate sentences in two separate paragraphs together. It also means that the readers won't know that there are other felonies besides murder, rape and robbery. If they know that, no one will be mislead.

As to misleading people, what about Tim only originally reporting one year of crime numbers to make his claim and no one but Tom on this section pointing out that you had to add it up over many years.

Tim, you have Lott on the brain. Here you have two co-authors and you have to write that "Lott made the claim."

Personally, I think the most cogent post in this thread is Matt McIrvin's regarding ChoicePoint. We should all read this carefully and follow up with a little research.Thank you sir.D

Dano - The guy who wrote the book on the Florida 2000 fraud is Greg Palast. Click on "theft of the presidency". Lots of other eye-opening reporting on Palast's site.

Bob - again, you are making arguments which are obviously irrelevant (blunder or dissembling - take your pick). Tim uses the 2000 numbers in order to get the breakdown of felonies to categories - not the absolute numbers. So the notion of summing felonies over the years is irrelevant.

By Pro bono mathe… (not verified) on 07 Mar 2005 #permalink

Pro bono, I hate to tell you this, but the percentages don't tell you anything by themselves about the Number of felons who can't vote. Either Tim is ignorant or dishonest or both. Your defense that you can some how figure out the total number of people who are felons without "summing felonies over the years" (or more accurately the number of convicted felons) is not serious math. In addition, the inability to define a simple word like guarantee properly is amazing. Even more disappointing, though not surprising any longer, is Tim's complete inability to admit a mistake like this.

Dear Tom, the number of felons who can't vote was given in the article. To find out the number of murderers, rapists and armed robbers who can't vote it is necessary to multiply that number by the percentage of felons that are murderers etc. That is what I did in my post. Everybody else seemed to be able to follow.
As for the total of murderers etc, I suggest you email Lott and ask for his evidence that the number is millions. How about it?

GIve me a break. In order to make this claim you have to ignore the fact that Lott&Glassman were making two different points in two different sentences in two different paragraphs. One is that all felons would be guaranteed the right to vote. The other is those felons who were currently prevented from voting. The 4.7 million only applies to the later, not the former which is where all your analysis was being applied.

Tom, you are the only one claiming that they were referring to all felons in that sentence. I don't think they were and neither do the other people commenting.

Tim, you can't even get this right. Bob agreed with me earlier. Please try stopping making things up. Especially when they are so easily checked.

Wow, these are strong responses Tim and "Dominion." You don't have an intellectual argument to respond to Tom so you start this silly stuff. Add me to one person who agrees with him.

Me am one person who agree with Tom, bob and Sam too.

So am I.

Er, do I need to point out that Tom used Bob to dispute Tim's claim that the only one that thinks Lott is referring to all felons is Tom? If Tom and Bob are indeed the same person (easy enough for the owner of a website to check)then this is not an honest answer, since what Tim states is true, the only one making the argument is Tom

Should I make an intelligent reply to a person that is simply agreeing with themselves? Why?

Do you know why being a sock puppeteer is a bad thing?

And I suppose while we are at it, I should state that I am one that agrees with Tim. Since that seems to be important to the "intellectual argument" set.

A delightful irony of the entire situation in the US is that prisons are increasingly built in rural areas to provide work for the locals. Since for census purposes, people in jails count, this swells the population of those areas. Since seats in legislatures are allocated by population and not voting population, well, now you know why rural areas are often over-represented. And, of course felons (especially incarcerated felons) cannot vote. Just peachy.

All arguments regarding statistics and definitions aside (let's all just agree that's its a large number), a fundamental issue has been raised:

Should convicted felons (or even the subset of convicted felons of violent crimes) be allowed the right to vote?

My position is no. As long as our criminal justice system allows for suspension of freedom (ergo incarceration), I see no reason why it is unreasonable to deny other rights. I see no slippery slope here.

Remember all rights granted by the consitution (even the first amenndment) are not unlimited. The argument of shouting "Fire" in a movie theater applies. So I see reason why the right of voting can't also be limited.

Kevon. no-one idsputes the legality (in the US) of laws limiting the right of (ex-)felons to vote.

However "can" and "should" are different.

By Ian Gould (not verified) on 13 Jul 2005 #permalink