Philosophy Discussion Forums | A Humans-Only Club for Open-Minded Discussion & Debate

[Beyondthecave]

RJG wrote,

if we wish to rationally determine if guns are good or bad, then we need to compare the good with the bad; the 'pros' with the 'cons'. ...for a one-sided singular view of the 'bad', ALWAYS looks bad!

I was specifically responding to your claim that high gun ownership rates in your town make all the people safer. I made no claim about whether, on balance, guns are bad. The evidence suggests that high gun ownership rates in your town indicate that, other things being equal, life in your town is probably less safe than it would be with lower gun ownership rates.

[RJG]

...life in your town is probably less safe than it would be with lower gun ownership rates.

Maybe yes, maybe no. Again, we can't know if it is "less safe" or "safer" until we compare the good with the bad. Just looking at the bad, tells us nothing of substance.

[Beyondthecave]

@RJG wrote,

Again, we can't know if it is "less safe" or "safer" until we compare the good with the bad. Just looking at the bad, tells us nothing of substance.

False. There are many good things and bad things that are unrelated to the inverse correlation between safety of populations and gun ownership rates within the populations. For example, what does ice cream have to do with that correlation?

By definition, a person is more safe if they are less likely to die or be injured.

Injury and death rates tend to correlate since they tend to have similar causes. Thus, death rates are a good indication of safety.

There are types of death that guns facilitate, like homicide and suicide, and others that are at best tangentially related to guns like death from cancer. Accordingly, comparing gun ownership rates to homicide death rates and suicide death rates is the reasonable method for determining the effect of guns ownership rates in populations on safety of those populations.

Guns are sometimes alleged to improve safety of populations by discouraging attacks. This is called "the deterrent effect". If the deterrent effect of gun ownership is the strongest factor, then that will show up in the data by giving us an inverse correlation between gun ownership and homicide death rates. More guns will result in fewer homicide deaths. But that is not what we find in the data. More gun ownership in populations generally correlates with a higher homicide death rate in those populations.

Of course, I am not saying gun ownership rates are the only factor that correlates with higher homicide and suicide death rates. However, the fact that other factors also correlate with high homicide and suicide death rates does not erase the fact that high gun ownership rates also correlate with high homicide and suicide death rates.

[RJG]

More gun ownership in populations generally correlates with a higher homicide death rate in those populations.

This (homicide deaths) is mostly irrelevant to the personal safety (safer or less safe) of the gun owner. In most cases, the homicide victim is NOT the one with the gun. The owner of the gun is still alive and 'safe' (...hopefully 'safely' in jail!). And maybe if this poor homicide victim had a gun to protect himself, he wouldn't be a homicide victim.

Again, if we are talking about the safety of the gun owner (and NOT of the 'NON-gunowner'), then we need to look at BOTH the good and bad of gun ownership. Counting the number of dead bodies (homicide victims) is irrelevant to determining the personal safety of the gun owner. Also, and again, you just can't count only the 'bad' side of the equation. For example, if you live in an area where it is necessary to own a gun to protect you and your family, then this should also factor into the 'safety' equation.

[LuckyR]

More gun ownership in populations generally correlates with a higher homicide death rate in those populations.

This (homicide deaths) is mostly irrelevant to the personal safety (safer or less safe) of the gun owner. In most cases, the homicide victim is NOT the one with the gun. The owner of the gun is still alive and 'safe' (...hopefully 'safely' in jail!). And maybe if this poor homicide victim had a gun to protect himself, he wouldn't be a homicide victim.

Again, if we are talking about the safety of the gun owner (and NOT of the 'NON-gunowner'), then we need to look at BOTH the good and bad of gun ownership. Counting the number of dead bodies (homicide victims) is irrelevant to determining the personal safety of the gun owner. Also, and again, you just can't count only the 'bad' side of the equation. For example, if you live in an area where it is necessary to own a gun to protect you and your family, then this should also factor into the 'safety' equation.

True. The major source of the decreased "safety" of gun ownership lies in increased lethality of suicide and gun accidents.

Feel free to mention that folks you know aren't particularly prone to one or both. Everyone knows that vulnerability to these are not evenly distributed within the population.

[Beyondthecave]

RJG, the point I am disputing is this claim you made.

But if you live in a small rural Texas town as I do, then 80% + of us own at least one gun, with all of us still alive, and without being a homicide victim. So gun ownership in our town keeps us all very safe (no one wants to mess with us that carry guns).

This was a claim about the safety of your whole community, not about the personal safety of an individual gun owner.

In your last post, for the second time, you argued a different question, albeit, the original question of this thread: Is it generally true that owning a gun makes a gun owner safer than he or she would be if he or she did not own a gun? You can argue that question with someone else if you like. I am not interested in discussing that issue with you. My dispute was with the claim you made that I quoted above. So far, you have failed to provide a persuasive defense of that claim. Is it defensible? I doubt it.

You claim no one has died in your community and that this is because of the deterrent effect of guns rather than a matter of luck. I point out that statistics show that the deterent effect of guns prevents fewer homicides than the number of homicides and suicides that correlate with high gun ownership rates die to the fact that guns facilitate homicide and suicide. Consequently, if your town has not experienced homicides or suicides, it is not likely due to high gun ownership rates. It is more likely due to luck or some other factor. Your theory that guns kept your town safe runs counter to the evidence that high gun ownership rates generally make populations of people less safe.

[RJG]

You claim no one has died in your community and that this is because of the deterrent effect of guns rather than a matter of luck. I point out that statistics show that the deterent effect of guns prevents fewer homicides than the number of homicides and suicides that correlate with high gun ownership rates die to the fact that guns facilitate homicide and suicide.

Do these statistics take in consideration the number of guns per capita (the ratio of guns-to-citizens in a particular community)?

I suspect that you, and the statistics, are probably correct in those 'low' gun-to-citizen ratios. In this group the 'majority' of those guns are probably in the hands of those they shouldn't be in (i.e the 'bad' guys; the thugs, gang members, etc), as the normal citizen (aka the 'good' guys) in this group don't typically pack a gun, thereby yielding a higher percentage of gun related crime/death per capita.

And in the those 'high' gun-to-citizen ratios (like my community where guns out number citizens by more than 2 to 1), then the majority of guns are in the hands of the responsible 'good' guy citizens, yielding a lower percentage of gun related crime/death per capita.

Consequently, if your town has not experienced homicides or suicides, it is not likely due to high gun ownership rates.

If you are talking about the 'percentage' (as opposed to the total number) of homicides/suicides, then I beg to differ. I definitely think there is a positive (decrease in the 'percentage' of homicides, but maybe not suicides) correlation with those communities that have high gun-to-citizen ratios.

It is more likely due to luck or some other factor. Your theory that guns kept your town safe runs counter to the evidence that high gun ownership rates generally make populations of people less safe.

If you are implying that the total 'number' of gun related deaths increase with an increase in the total 'number' of guns per capita in the community, then I can't disagree.

But if you are implying that the total 'percentage' of gun related deaths increase with an increase in the total 'number' of guns per capita in the community, then I can certainly disagree.

[Beyondthecave]

RJG wrote:

I suspect that you, and the statistics, are probably correct in those 'low' gun-to-citizen ratios. In this group the 'majority' of those guns are probably in the hands of those they shouldn't be in (i.e the 'bad' guys; the thugs, gang members, etc), as the normal citizen (aka the 'good' guys) in this group don't typically pack a gun, thereby yielding a higher percentage of gun related crime/death per capita.

And in the those 'high' gun-to-citizen ratios (like my community where guns out number citizens by more than 2 to 1), then the majority of guns are in the hands of the responsible 'good' guy citizens, yielding a lower percentage of gun related crime/death per capita.

OK. That is your hypothesis: More guns per capita results in fewer homicide deaths. Rational people proceed to test their hypotheses.

If states with high guns per capita have lower homicide death rates, your hypothesis will be supported. But, it states with low guns per capita have lower homicide death rates, then your hypothesis is disconfirmed.

The best evidence I have been able to find is not on guns per capita, but on the percent of state residents that own guns. I will use that as a stand in for guns per capita. It seems to me that the percentage of gun owners is the more relevant statistic to use any way. If in one community, each person owns one gun, but in another community, one person owns twice as many guns as there are people in his community, the second community will have more guns per capita, but your argument that lots of gun ownership is good because good people with guns will outnumber bad people with guns works better for the community where each person owns one gun. Your argument would apply to communities with a higher percentage of gun owners rather than higher guns per capita rates (in those cases where the two statistics differ).

The data on percentage of gun ownership in each state that follows is from a scholarly journal, Kalesan B,et al., Injury Prevention,2015;0:1–5. doi:10.1136/injuryprev-2015-04158. https://injuryprevention.bmj.com/conten ... x0laFZMsQ2

The data is based on interviews of a representative sample of 4000 persons.

The five states with the highest percentage of gun owners is listed below in order. The first number after each state is the percent of residents who are gun owners. The second number is the number of homicide deaths in that state per 100,000 residents. The Homicide death rate data is from the US Center for Disease Control. https://www.cdc.gov/nchs/pressroom/sosm ... micide.htm

Alaska 61.7% 10.6
Arkansas 57.9% 9.8
Idaho 56.9% 3.0
West Virginia 54.2% 6.5
Wyoming 53.8% 3.4

The average homicide death rate for these high gun ownership states is 6.7 homicide deaths per 100,000 residents.

Now, lets look at the five states with the lowest percentage of gun owners in order:

Delaware 5.2% 6.9
Rhode Island 5.8% 1.8
New York 10.3% 3.0
New Jersey 11.3% 4.1
New Hampshire 14.4% 1.3

(In the case of Wyoming, New Hampshire, and Rhode Island, the CDC listed the death rates as zero because of the smaller sample size. I computed the actual death rates based on the homicide deaths reported by CDC and the populations of those states as provided by the US Census.)

Clearly, these statistics show that homicide death rates tend to be lower in states with lower rates of gun ownership. Of the five low gun ownership states, only Delaware exceeded the average of the high gun ownership states, and only by a small margin.

The average homicide death rate for the low gun ownership states is 3.4 homicide deaths per 100,000 residents. That is approximately one half of the homicide death rate in the high gun ownership states.

Conclusion: Your hypothesis is wrong.

Explanation: I fully understand that gun ownership can have a deterent effect. A would be killer might act differently if he fears he will be shot by the good guy with a gun. However, there are other phenomena that are relevant. Many killings are impulsive. People get mad or frightened and they act on impulse. Guns make those impulsive attacks easier to carry out and make them more lethal. Guns are killing machines. That is their function and they are well designed for that purpose. What the homicide statistics show is that guns facilitate homicides more often than they prevent them, even in states with high gun ownership rates.

[LuckyR]

Not so much. Since the acquisition of the guns is not random, no one can say that guns cause the murders or if states with high murders make folks nervous to the point of purchasing more guns

[RJG]

Since the acquisition of the guns is not random, no one can say that guns cause the murders or if states with high murders make folks nervous to the point of purchasing more guns.

Good point.

What the homicide statistics show is that guns facilitate homicides more often than they prevent them, even in states with high gun ownership rates.

But this hypothesis is easily proven FALSE, by the 'reality' of those small rural towns that have very high gun ownership -to- very low homicide rates.

Remember, statistics only show us a 'correlation', it doesn't show actual 'causal results'. Who knows, maybe the actual cause of high homicide deaths is NOT the 'number of guns', but instead the 'number of thugs'? ...since guns can't kill by themselves, this seems reasonable, right?

So if we generate statistics comparing 'thugs' -to- 'homicides', then this may be a more accurate assessment of the causal result of homicides. If we have no way to count the number of 'thugs', then maybe look at "unemployment rate", or "education level", (...or some other possible attribute of 'thugness', ...sorry, no offense to all you 'thugs' out there.)

Closing our eyes to the statistics surrounding the 'types of users' of guns, because one is "hell-bent" on demonizing 'guns', is not a fair, nor an honest evaluation of the situation.

[Eckhart Aurelius Hughes]

The question is about the sum statistical net effect (i.e. the net consequential result when all the ways gun ownership make the owner safer are weighed statistically with all the ways gun ownership makes the owner less safe).

Sometimes access to firearms saves the owner's life; sometimes access to firearms does the opposite. Nobody is denying it does both. That was a given, explicitly stated long ago in this topic. The titular question is simply regarding the total net effect.

Yes, agreed! But, to calculate the "total net effect", you also need to know how many were 'saved or non-injured' because of guns. You are only accounting for the people killed/injured, and neglecting those saved/not-injured.

You cannot calculate "net effect" with only one variable (or one-side of the equation).

I am not calculating only one side. I am not asking people to calculate only one side. I have no idea how a person even could calculate one side.

I am not only accounting the people killed/injured by guns.

Rather, we are including in the calculation all four of the following:

(1) non-gunowners who were killed or injured (e.g. those who may have been killed or injured because they didn't have guns in the sense that they would have been saved from their death or injury if they had a gun)
(2) non-gunowners who were not killed or injured
(3) gunowners who were killed or injured
(4) gunowners who were not killed or injured (e.g. who may have been saved from death or injury as a result of their gun ownership)

#1 + #2 = 100% of non-gunowners

#3 + #4 = 100% of gunowners

#1 + #2 + #3 + #4 = 100% of people

When I ask for the net effect, it necessarily entails calculating all four of those.

What do "non-gunowners" have to do with anything??? --- What do bananas (non-apples) have to with the 'color' (green or redness) of apples?!!! (i.e. "What do non-gunowners have to do with the 'safety' (safer or less safe) of gunowners?")

If we wanted to find out whether eating green foods is correlated to being healthier or less healthy, then you would have to statistically compare the average health of people who eat green foods AND of people who do not eat green foods. Of course non-green-fruit-eaters would have something to do with anything in trying to find out whether eating green fruit is correlated with better health (or anything). (Your seeming rebuttals are at best verging on the utterly absurd.)

Correlation is found through the comparison between the groups.

That is why one statistic always fails to provide a correlation. For instance, people might make the fallacious mistake of saying just because X% of people who do hard drugs started by first smoking marijuana that therefore marijuana is correlated to hard drug use (and may be a gateway drug). That is utterly fallacious. Such a conclusion requires additional to compare to the X% and see if they are higher or lower. Even if it only assumes correlation not causality, it is still utterly fallacious. Correlation is found through the comparison of multiple stats. Indeed, it might be true that 90% of hard drug users drank their mother's milk as babies too. One stat alone can never prove correlation. You MUST one way or another compare the stats of effects of gunownership AND effects of non-gunownership to find what correlates with gunownership.

It's a comparison between the statistics of those don't own or do something to those who do that is the correlation by definition.

Whether or not gunownership is correlated to one being less safe or more safe is by definition a comparison of the average safety of non-gunowners compared to the average safety of gunownership.

As you said in your own words, "You cannot calculate 'net effect' with only one variable (or one-side of the equation)." But you are the only in this topic suggesting doing such an absurd thing.

What you are suggesting (not looking safety of nongunownership) is utterly fallacious. The meta-analysis in question did NOT commit that fallacy.

You accused the meta-analysis (and/or me) of committing that fallacy of only looking at one side. No, the meta-analysis doesn't. You are the one proposing only looking at one side, which is a fallacious. Correlations are by definition comparisons between both sides.

Since no counter-evidence to the meta-analysis in question has been provided, my tentative conclusion is still that (like automobile ownership) gun ownership is correlated with the gun owner being less safe (i.e. more likely to get injured or killed than those don't own guns). Should equal or better counter-evidence be provided, I will update my tentative conclusion according to the evidence.

[RJG]

What do "non-gunowners" have to do with anything??? --- What do 'bananas' have to with the 'color' of apples?!!!

If we wanted to find out whether eating green foods is correlated to [who?] being healthier or less healthy…

Here you are seemingly referring to 'anyone/everyone' (both green food eaters AND non-green food eaters).

Does gun ownership make the GUNOWNER safer or less safe?

And here you are specifically referring to the 'gunowner' safety (and NOT to anyone else's safety).

If you are concerned about the 'color' of 'apples', then discussing bananas (non-apples) is a non sequitur. -- and likewise, if you are concerned about the 'safety' of 'gunowners', then discussing 'non-gunowners' is a non sequitur.

Correlation is found through the comparison between the groups.

Yes. But again, if you are interested in the 'safety' or 'non-safety' of the gunowner, then you are comparing the WRONG groups. You need to compare the 'safe' aspects (of owning a gun) to the 'non-safe' aspects (of owning a gun). The group of 'non-gunowners' have no business being compared to here.

[RJG]

Correlation is found through the comparison between the groups.

Comparing the safety of gunowners to the group of non-gunowners is absolutely nonsensical. It is like comparing apples-to-oranges. It is like comparing the group of apple eaters with the group of non-apple eaters ('orange' eaters), so as to then non-sensically determine the safety of apple eating.

If one-out-of-a-billion apple eaters chokes and dies on the core of an apple, then the death rate of eating apples will always be greater than the death rate of non apple eaters (who have ZERO deaths associated with eating apples! ...because they don't eat apples!!!)

If you wish to know if eating apples are "safer or less safe", then you need to compare the 'good' (safe) with the 'bad' (unsafe) aspects of eating apples. Non-apple eaters have no business being compared to here!

[RJG]

Does gun ownership make the gunowner safer or less safe?

Does apple eating make the apple-eater safer or less safe?

-- How would we determine the answer to either?

[Eckhart Aurelius Hughes]

Correlation is found through the comparison between the groups.

Comparing the safety of gunowners to the group of non-gunowners is absolutely nonsensical. It is like comparing apples-to-oranges.

That's like saying that comparing the health/safety of smokers to non-smokers is absolutely nonsensical. No, it's basic correlation statistics, and calling it nonsensical is nonsensical. Trying to determine whether smoking cigarettes is healthy/safe or not without comparing the average death/illness rate of smokers to the average illness/death rate of non-smokers is absolute nonsense. What you are suggesting is nonsense.

It is like comparing the group of apple eaters with the group of non-apple eaters [... to] determine the safety of apple eating.

That is exactly how to study the healthfulness of eating apples. Your example only further proves my point.

If one-out-of-a-billion apple eaters chokes and dies on the core of an apple, then the death rate of eating apples will always be greater than the death rate of non apple eaters (who have ZERO deaths associated with eating apples! ...because they don't eat apples!!!)

Nobody is suggesting only calculating violent deaths caused by guns (let alone only calculating those involving the gun owner's own gun). Rather we are calculating all violent deaths, including those done without guns at all and those done by an assailant with the assailant's own gun.

If you wish to know if eating apples are "safer or less safe", then you need to compare the 'good' (safe) with the 'bad' (unsafe) aspects of eating apples. Non-apple eaters have no business being compared to here!

That is absurd. Of course, you would compare the average health outcomes of non-apple-eaters to the average health outcomes of apple-eaters to determine the statistical health effects of eating apples. Your suggestion otherwise is extremely absurd.

Does gun ownership make the gunowner safer or less safe?

Does apple eating make the apple-eater safer or less safe?

-- How would we determine the answer to either?

To find the correlation, you simply check:

#1 - The percentage of people who eat apples who get hurt/sick during a certain time frame
#2 - The percentage of people who do not eat apples who get hurt/sick during a certain time frame

If #1 is greater than #2 than eating apples is correlated to getting hurt/suck more often (i.e apple eating is correlated with being less safe), if #2 is greater than apple eating is correlated to getting hurt/suck less often (i.e. apple eating is correlated with being more safe), if #1 is roughly equal to #2 then there is no significant statistical correlation between apples and getting hurt/sick.