Gun Violence and Crime in the US

Gun violence, or gun crime, in the US seems to be a very messy problem.

What I aim to do with this post is discard all of the noise and illuminate the facts. I have drawn heavily on robust statistics and studies and progress through them in a logical manner so that we can have a clear understanding of the actual problem and how it occurs.

I compare these statistics to what is known about human combative behavior. I then compare various arguments to the above in order to present gun violence in the US as it is.

This post is all about gaining a clear appreciation of gun crime in the US. Although this post is focused on the US it is important for self defence[1] thinkers from all modern countries to understand how gun crime occurs in the US as there are fundamentals that apply to all violent crime and specifically sub sets such as gun crime.

Throughout this whole post we will be looking through the lense of a self defence thinker. For Low Tech Combat readers this post is all about understanding gun violence and gun crime. It is only when understanding has occured that a strategy or approach can be developed to lower the risk of becoming a victim of gun violence.

Framing the problem

There are many views on gun violence and gun crime in the US. We will get into these, and the facts, but before we do I would like to frame the whole gun violence problem with a quote from Sam Harris:

How is it that we live in a society in which one of the most compelling interests is gun ownership? Where is the science lobby? The safe food lobby? Where is the get-the-Chinese-lead-paint-out-of-our-kids’-toys lobby? When viewed from any other civilized society on earth, the primacy of guns in American life seems to be a symptom of collective psychosis.

Whilst it may seem like Mr Harris is against guns we can balance our entry to this topic with an alternate quote again from Sam Harris::

A world without guns, therefore, is one in which the advantages of youth, size, strength, aggression, and sheer numbers are almost always decisive. Who could be nostalgic for such a world?

Both of these quotes illustrate aspects of the two sides of the debate and are just a teaser into the ongoing gun debate. We will get more into this aspect further into this piece.

With a somewhat balanced overall view we can now begin to explore this topic with open eyes and an open mind.

Gun ownership in the US

Before we start delving down into the gun crime nitty gritty it is probably worth asking the following:

Question: How many guns are in the possession of US citizens?

Answer: 286 Million firearms!

Between the United Nations of Drug Control (UNODC), Pew, Gallop and Harris Interactive it has been determined that for every 100 Americans there are 88.8 firearms. ~322 Million people live in the U.S which gives us the figure of 286 Million firearms. This is the highest per capita, as well as total numbers, in the world. Yemen comes in second with just a paltry 54.8 firearms per 100 persons. Lame hey?

The below image showcases the above points quite well[2].

armed-to-the-teeth

In order to generate some perspective of the gun violence problem in the US right now it is useful to zoom out and look at violent crime rates in the US over the last 52 years so that we can better appreciate where we are today.

52 Year Violent Crime Trends

52-year-violent-crime-us

The above chart shows overall violent crime rates in the US over a 52 year time frame (not just guns). I have used a very large data set to populate these charts which is the only one I know of that holds this sort of data over this time frame[3]. You can visit a previous post where I dig into the 50 year data set in a different way here.

It is worth commenting, for those of you who have never seen this data before, on the significant trend reversal in the 1990’s. Violent crime significantly began to drop in numbers around this time.

Nobody really knows for sure how or why this occurred. There are some theories which cover both the carrot and the stick. Briefly, some of the theories include the following:

  • Job growth - lowering unemployment.
  • Increase in arrests.
  • War.
  • Increase in eduction for high crime communities.
  • Removel of lead from gasoline.
  • Ageing of baby boomers.
  • Harsher penalties for violent crime.
  • Legalised abortion.
  • End of crack epidemic.

I won’t go into the pros and cons of each of the above at this time. We will move on and stick to the topic at hand.

Now that we have a broad appreciation of where violent crime rates have been in the past and how they have trended and dropped off in the 90’s we will now compare gun crime rates alongside the overall violent crime rates.

Gun crime rates

gun-crime-rates

I cannot find data that goes back over 50 years for gun crime. The best data I could find is from the US Bureau of Justice Statistics which covers the time period between 1993 and 2011[4]. I overlaid the gun crime data onto the overall violent crime chart for comparison purposes.

Note on Statistics Sources: The BJS (National Crime Victimisation Survey or NCVS more specifically), uses victim surveys of incidents that may or may not have had police involvement. Whereas the FBI UCR tool relates to incidents that have gone through police reporting channels. So the NCVS numbers can be higher than those found in the UCR tool for various segments of data.

From the chart above we can see that gun crime generally mirrored overall violent crime over that same period. The limited gun crime data captures the drop off from the 90’s quite well. The drop off in gun crime was actually more rapid than more general violent crime.

The gun crime data also generally matches the slight rise in violent crime numbers again around 2006. So we can see that gun crime generally matches changes in violent crime.

Whatever causes changes in violent crime also changes the rate of gun crime. I can see no indication that gun crime goes against the grain in this regard.

What is interesting is that gun crime made up a larger percentage of overall violent crime in the 90’s. Although we won’t go into the percentages too much (due to differences in data sourcing), we can see that gun crime makes up less of a portion of overall violent crime now than it did in the 90s. So we can expect less of a chance of encountering gun crime now than in the 90s. However, this risk is still far higher in the US than most other modern countries.

Gun Crime

Let’s explore what types of incidents gun crime entails. The categories that can be broken down underneath the umbrella term of ‘gun crime’ are as follows:

  • Aggrevated assault
  • Armed robbery
  • Homicide
  • Non-fatal violence[5]

We will now look at how these various categories make up total gun crime in the US

gun-crime-percentages

Obviously homicide sees large percentages of gun use. Non-fatal violence is clearly at the bottom. The interesting part is the closeness of aggrevated assault and armed robbery. Due to research I have conducted previously I feel confident in providing some commentary regarding these two types of violent crime and some of the data points to follow. I have identified some nuances that I will draw out as we go on.

stranger-violence

Typically, aggrevated assault comes about due to some form of argument or disagreement often in the presence of some type of crowd. This sort of crime is generally driven by status and ‘face’. Alcohol is frequently involved.

This violence involves people who are battling over their status within a social group. This is what we can see in the chart of non firearm violent crime. We are seeing more violence between people who know each other.

It can also occur between people from different social groups. This is how many pub fights start. It is the same sort of violence. It starts off and happens the same way although it can often relate to status of one group compared to another as opposed to between people.

There was much less shooting between people who know each other than between strangers. This is natural. It makes no evolutionary sense to kill off people from within your own social group. It is easier to shoot a person who you do not know. You do not know if they have children, you have never spoken to their mother and you cannot empathise with their perspective etc.

Non firearm violent crime happens more with people who know each other. Non lethal firearm violence happens more between strangers.

But note the percentages are not 90/10. It is much closer than that and certainly not clear cut.

gun-crime-locations

This chart provides some further insight into how gun crime occurs in the US. We know from the previous two pie charts that gun crime sees most shootings occur between strangers with about 40% occuring between people who knew each other.

This chart shows where this sort of gun crime is occuring. It is occuring in and around people’s homes. In fact 45% of all gun crime happens in and around homes. This is the largest segment. This makes sense but it is interesting to see this reflected in the data. The amount of gun crime occuring in and around homes closely matches the rate of gun crime occuring between people who know each other.

What I am not saying is that all gun crime happening in and around homes is between people who know each other. That is jumping to conclusions. Some of that crime will be property crime gone bad or otherwise known as home invasions which go bad.

If we look at overall homicide rates involving strangers, not just gun related, between 1993 and 2008 - Source BJS[6], we can see that the two biggest circumstances leading to the homicide were robbery (19.3%) and an argument (23.8% - not over money which was just 1.1%). These were the two clear leaders. Of note, burglary saw just 1.3%.

So the traditional fear of the stranger kicking our door in during a savage home invasion is a very small risk. There are other far more significant risks to concern ourselves with if we are actually focused on understanding violence for self defence purposes and not on justifying an agenda.

Furthermore, according to the same BJS report, between 2005–2010, just 9% of violent victimisation that occured in victims homes was committed by strangers. Now these numbers are not specifically regarding gun crime but that low number really speaks for itself. The vast majority of violent crime happening in and around homes is between people who know each other and not from scary strangers.

The most likely places you are to be attacked with a gun by someone you do not know is in an open area, on a street or on public transportation. The next most likely places are a parking lot or garage.

These are the sorts of locations where Predatory crime occurs. These are the sorts of places where muggings occur. This is pre-meditated violent crime where offenders can choose to arm themselves before they go out on their hunt. This sort of crime is not about status. This crime happens away from crowds.

This sort of predatory crime is generally all about taking some sort of resource like cash, cards or smart phone from a suitable victim they have carefully selected. These sorts of crimes are typically much more lethal than aggrevated assault.

What weapons are being used to commit violent crime?

guns-in-homicide

In this chart we can see the overall drop off in crime numbers in handgun use. We don’t actually see too much of a change overall for all other types of firearms.

All other types of firearms are just that; rifles, shotguns and all other types. We can clearly see that handguns have been the most common firearm used in homicides over many years though less so now than in the 1990s.

Screen Shot 2015-10-06 at 8.50.36 pm.png

This chart details numbers of non fatal firearm incidents. First of all, in this chart we see a much more significant drop off in handgun related crime than in homicides.

Do note though the overall numbers are much higher than homicides. If you look at the 2011 segmants of both of the above charts you will be able to see that non fatal incidents of firearm use have a much higher percentage of handguns. This makes sense as any incidents involving a rifle or shotgun are more likely to result in death than a pistol.

However, even considering that we can see that handguns kill far more people than all other types of firearm combined. The following chart showcases this.

guns-in-homicide-2011

If we look at the most recent year in this data set (2011) and do the numbers we can see that handguns were used in 88% of all firearm related incidents. This chart makes the case that we need to be more concerned about handguns than all other weapons combined.

However if we only look at incidents that were lethal the numbers look a little different.

guns-homicide-2011

Here we can see that handguns make up 73% of all homicides. All other firearms are more significant in this category being responsible for almost a third of all lethal incidents. So this number is not huge but is significantly more than what is seen in non-lethal encounters.

It is worth mentioning however more lives can be saved by focusing on handguns, especially for self defence purposes.

Firearms used in self defence?

Often pro gun groups like to highlight all of the cases where guns have been used in self defence. This seems fair as everyone has a right to self defence. It would be shameful to disarm people who might later become victims of violent crime.

Self defence is a big theme often used to support the notion that we should all arm ourselves to defend against the violent criminals out there. I have heard on more than one occassion a number of 2.5million instances of guns being used in self defence per year. At first glance this seems a large number.

But what do robust statistics tell us? Let’s have a look. The NCVS have just this sort of data - published via BJS. According to this data set, between 2007 and 2011, there were over 29.5 million instances of violent crime in the US.

Let’s have a look at the below chart for some interesting insights.

gun-self-defence

Even without looking at this chart for insights into gun crime I find it fascinating to see what people have generally done in self defence situations. Interestingly almost half (44%) offered no resistance. Offering no resistance can be a very smart thing to do when being mugged. Give the predator what they want, the resource, and they will generally leave things at that and quickly leave the scene. I hope that not too many of these incidents involved ‘adult bully’ situations where an aggressor was picking on an easy victim to inflate their ego. I guess these victims are part of the group that seek to join martial arts classes.

A quarter (26%) used non confrontation tactics which is great to see. This was a higher figure than I thought it would be but I would like to see that figure up past 75%! Another quarter (22%) threatened or attacked without having a weapon - so unarmed.

Then we see the figures drop off dramatically. If you squint and look really carefully you will see that just 0.8% of all violent crime encounters over the reporting period saw people defending themselves with a firearm. Over the 5 years that is just 235,700 cases out of 29.5 Million cases of violent crime. Hardly the 2.5 Million cases per year as bandied about by some elements. Even statistical errors could not account for that discrepancy.

This 0.8% figure is especially low when we consider that in the US there are 88.8 firearms per 100 people. An obvious assumption is that the guns that are in the US are not being carried or used for self defence purposes. They are being used for something else. There is certainly no case to argue that there should be more guns.

Slightly more recent data has seen the Violence Policy Center release a study, published in June 2015, which captured data over a five year period (2008–2012). It found that for every 1 case of justifiable homicide (self defence using a gun) there were 38 criminal homicides.

The study expands on the above by identifying of those justifiable homicides 32.9% of people killed were known to the shooter[7].

Gun ownership

One common argument is that higher levels of gun ownership lowers violent crime. After all who would want to commit a crime if the victim was likely armed with a gun?

This sort of belief was confirmed in a recent Gallup poll which demonstrated 63% of Americans believe having a gun in the house makes it safer. This figure has nearly doubled since the year 2000.

Unfortunately this belief, though common, is inaccurate. A recent study published in the ‘American Journal of Preventive Medicine’ shows us that increased gun ownership levels saw increased rates of firearm assault and firearm robbery. There is also a significant association between firearm ownership and firearm homicide.

Another detailed study published by the US National Library of Medicine, National Health Institute, titled ‘Gun ownership and firearm-related deaths’. This study looked at 27 developed nations and produced results indicating a very strong correlation between guns per capita per country and the rate of firearm related deaths.

There does seem to be one anomaly here worth noting. Adam Winkler (a Law Professor at UCLA) in his book Gunfight: The battle over the right to bear arms (Amazon link) shows that concealed carry permit holders are involved in less violent crime than not only the average citizen but less than police officers.

Gun ownership on rise?

gallup-poll-guns

As per the above chart, a gallup poll indicates that guns are present in 42% of homes. Just 3 years earlier the figure was 45% which at the time was the highest levels since 1993 (51%). Although do note there have been periods over that time between 1993 and 2011 that approached 43% so that 2011 figure was not coming off a major low period.

The Problem ‘Number Crunched’

When we factor in that there are 286 Million firearms in the US, 135 Million Americans own a firearm and roughly 5.9 Million violent crimes are occurring a year it is really odd that we are seeing figures like only 235,700 cases of firearms being used in self defence over a 5 year period. That’s just 47,140 cases of firearms being used in self defence a year. That deserves its own quote.

286 Million firearms in the US. ~5.9 Million violent crimes per year. Only 47,140 instances of guns being used for self defence per year.

There seem to be plenty of guns; they just don’t seem to be getting carried and used for self defence purposes.

The US Gun Violence problem summarised

Here are the key points we have learnt from a self defence perspective.

  • There are hundreds of millions of firearms in the US.
  • The US has the highest per capita, and overall numbers, of guns in the world.
  • Gun crime rates generally mirror those of general violent crime.
  • Gun crime is about a third of what it was in the 1990s.
  • ~70% of homicides involve a firearm.
  • Most firearm violence occurs between people who do not know each other (but not by much)
  • Almost half of all gun crime occurs in and around people’s homes and the vast majority of it is between people who know each other.
  • The most likely location you are to encounter a Predator with a firearm is on the street, open land, parking lot or public transport.
  • By far, the most likely weapon you are to encounter is a pistol (or handgun).
  • The two biggest incidents leading to homicide are robbery and an argument (these are total homicide statistics but when we consider ~70% of all homicides involve a firearm so it is definately worth noting).

Self Defence approaches for gun defence?

All standing measures for lowering risk of encountering violent crime are relevant. There are few approaches that need to be taken specifically for gun defence. Long time Low Tech Combat readers will have recognised that few of the gun crime statistics surprise.

Some general truths that remain consistent are:

  • More lethal encounters occur between people who do not know each other.
  • Predatory crime occurs on the street and other open spaces.
  • Assault or violence involving people who know each other is far more common than crime involving strangers.
  • Crime between people who know each other is generally less lethal.

Some aspects unique to guns worth noting.

  • An encounter, unarmed, that would have resulted in a black eye or ripped hair can instantly turn lethal when a gun is involved.
  • When you carry a gun you can take a view that you are carrying a hammer and everything starts looking like a nail. Meaning you may be quick to draw it without considering other options.
  • To carry a gun safely and responsibly requires significant training. To deploy it safely and effectively in a real self defence situation requires even more specialist training.
  • Considering the above, a criminal carrying a gun is likely to see everything as a hammer and not care about changing that approach.
  • When considering defences for muggings etc, you MUST include gun defence as part of your training (and against three or more attackers).
  • When getting into an argument with a friend, family member or casual aquaintance there is a very real possibility that person may be armed.

I will state it again here - all knowledge aspects, avoidance and de-escalation approaches apply for gun defence too. Do not get fixated on thinking about gun defence as just about disarms and quick draws. Truly smart self defence occurs before physical encounters happen.

What to do about gun crime in the US?

Most people reading this are probably feeling I am about to talk about gun control.

I am not.

I think there are too many guns in US for that to be effective. I think it would be practically impossible to get guns out of the hands of criminals.

Gun control has lowered gun crime in other places of the world but I just think the problem is too big in the US and will generate a high level of public unrest and instability that may not be worth the cost. Countries where gun control has worked have had far less popular resistance and much fewer guns per capita in the first place.

Besides, gun crime is just a symptom of the real problem. The real problem is that there are Americans who want to kill other Americans; from school kids to adults. There are significant social problems at play here. This is the messy part about gun crime.

From a self defence perspective understanding how, where and why guns are used to commit violent crime, as we have seen, is fairly achievable. Once we understand how gun violence occurs we are in a much better place to consider how to defend against it.

However I am not about to get into this messy social aspect here. There are a range of options out there to lower gun crime levels in the US that seem quite pragmatic and sensible however seeing this site is about self defence it is really a topic for another time and probably another place.

Below is a convenient list of just the key sources cited in this post:

Feel free to leave your thoughts below.

Header image source raymondclarkimages

All charts and graphs by Adam at Low Tech Combat unless otherwise stated.


  1. Yes self defence is spelt that way on purpse. I am from Australia and we write self defence the way it has always been in the English language - a ‘c’ instead of an ‘s’. Americans seem to get confused by my spelling as they changed a few english words around to show they no longer need the British empire and can come up with their own interpretation (note my joking tone may not translate through just words). ‘Center’ is another one - we still spell it ‘centre’. But I digress…  ↩

  2. I got this data and chart from http://www.huffingtonpost.com/2012/08/09/gun-ownership-infographic_n_1762059.html. I also conducted cross checks with Gallop and the like and they all have similar data so I consider this combined assessment as an accurate representation of the situation on the ground.  ↩

  3. We have used the FBI Uniform Crime Reporting (UCR) data tool found here, http://www.ucrdatatool.gov. This dataset is police reported crime. So this is crime that involves police being involved in an incident. This compares to the NCVS data that we use later which is generated from surveying victims of incidents that may or may not have had police involvement. Hence the NCVS numbers of violent crime are higher than the UCR numbers. However the benefit of the UCR data is that it goes back so far. Much further than the NCVS data. This shows us longer trend patterns.  ↩

  4. The BJS link provides access to raw data obtained by the NCVS and Centers for Disease Controls (CDC) National Vital Statistics System. This data can be skinned in many ways. It contains data primarily focused around gun crime over a number of years - typically around an 18 year period that is then crunched into a range of useful ways such as location of crime, whether the attacker knew the shooter or what the person getting attacked did in self defence. Very interesting stuff. To get total gun crime numbers I combined the gun homicides with the non fatal gun incidents. You can download a whole bunch of CSV spreadsheets filled with interesting data and place them all in an Excel, or Numbers, workbook and go nuts looking at all of that data. If you are not big on CSV they also conveniantly have a PDF for you. I had great fun going over all of this data and I am sure you can too. Now stop reading these epic footnotes and get researching!  ↩

  5. Nonfatal violence includes rape, sexual assault, robbery, aggravated and simple assault. A small percentage of rape and sexual assaults involved firearms but are not included in this data due to small sample sizes.  ↩

  6. Looking at general homicide rates is a fairly good indication of homicides caused by guns as gun violence causes 70% of all homicides Source.  ↩

  7. 56% of those shot were not known to the shooter and 11.2% of relations were unknown.  ↩