Crime as a Cause of Crime? Evaluating Routine Activities

Click to download pdf version

Felson and Cohen’s Routine Activities approach (1979) has arguably been one of the most-influential recent theories of crime, one that sits squarely within contemporary New Right / Realist explanations for crime and deviance.

This post looks at a couple of useful ways students can evaluate the approach.

A Quick Outline…

The main objective of this post is to introduce a-level sociology students to some points of evaluation for Routine Activities and in order to do this, students need to be familiar with the basic approach. If they’re not, this quick outline should bring them up-to-speed.

One advantage of the approach is that it’s easy to describe, at least in Felson and Cohen’s (1979) original argument that crime involves three essential elements:

The Predatory Triangle

1. A Suitable Target

2. A Motivated (latterly, a Capable or Likely) Offender

3. The absence of a Capable Guardian.

This formulation has, over the years, seen various additions (see, for example, Clarke and Eck, 2003) although whether these represent an attempt to refine the approach or merely a papering over of the theoretical cracks, I couldn’t possibly say), but one of the most significant refinements has been to express these elements as part of what Felson and Cohen call “the Predatory Triangle”.

While it might not seem much, this visual representation shows how the three elements are linked, such that, as they argue, a “Lack of any one of these elements is sufficient to prevent the successful completion of a direct-contact predatory crime”.

In other words, for a crime to take place a likely offender needs to meet a suitable target in the absence of a capable guardian. Or, as Felson (1998) puts it, by way of an example:

“A burglary needs a capable and motivated offender to find a suitable and accessible target, in the absence of anyone or anything there to keep the two apart”.

There are a couple of points we can note in this respect:

1. A crime occurs (or, in a weaker formulation, “is likely to occur”) when these three elements meet in time and space.

If you want to give Routine Activities an historical context, in common with other Right Realist approaches such as Broken Windows it’s based on Ecological (how behaviour is influenced by social and physical spaces) and Control theories of behaviour. An element of Rational Choice Theory is often thrown into contemporary versions of the approach to “explain” why offenders are motivated or demotivated in relation to criminal intent.

2. “Routine Activity” is an important part of the overall crime equation because it refers to the idea that “motivated offenders” come to identify suitable targets – and the extent to which they are protected by “capable guardians” – through their everyday experiences.

As with most New Right / Right Realist approaches, the main focus is on crime prevention. While there is necessarily some discussion / speculation about “criminal motives” – Felson and Cohen refer, for example, to “structural conditions” that produce “motivated offenders” – there isn’t a great deal of interest in trying to explain why people are motivated towards crime.

Crime, in this respect, is seen as a more-or-less unfortunate by-product of social and / or psychological background (depending on the particular theorist); crime is a “fact of life” and the best a society can do is to control it.

In this respect the basic “Predatory Triangle” has clear implications for the control of crime:

1. Motivated offenders can be deterred.

2. Suitable targets need to be “hardened” or protected in ways that make them seem unsuitable to offenders.

3. Capable Guardians – from the police, through careful and well-informed populations to various forms of technological guardianship (such as CCTV) – need to be extensively deployed.

Evaluating RAT

We can evaluate the Routine Activities approach in two main ways:

1. Theory?

Although Felson and Cohen (1979) avoided using the word “theory” (this seems to have been added at some later point) to initially describe their Routine Activities approach, it’s clear the intention was – and continues to be with the various iterations of Routine Activities – to produce a theory of crime prevention / control. It is, in this respect, an approach that seeks to initially model human behaviour (the Predatory Triangle) and use that as the basis for predicting the conditions under which a crime will or will not take place. As Sutton (2018) summarises the argument:

“The meeting in time and space of a motivated offender and a suitable crime target in the absence of a capable guardian (my emphasis) will inevitably lead to a crime occurring”.

Similarly, Robinson (1998) argues “If we can identify which potential targets in any given environment are most suitable for victimization, then we can design and implement crime prevention strategies to make them less suitable. This will prevent criminal victimization” (my emphasis).

The evaluative question here, however, is the extent to which Routine Activities can be considered a theory – and why this might be important.

A theory, in the sociological sense, has two important features that we can examine in turn:

1. It has predictive power: in this instance, the combination of motivated offender, suitable target and lack of capable guardian will result in a crime taking place.

As we’ve just suggested, Routine Activities does claim the ability to predict “predatory forms of crime”. The argument advanced by writers like Sutton, however, is that it only works by “predicting the past”, not future behaviour. Think about it in these terms:

  • We can only identify a “motivated offender” after the fact of their offending. Until they actually offend, they are simply “a potential offender” (or “person”).
  • We can only know if a target is “suitable” after it has been attacked (by a “motivated offender”). Until that point, it’s effectively an “unsuitable target” or, indeed, not a target at all.
  • We can only know if a guardian is “capable” until the moment it’s not. That is, the “capable guardian” fails to prevent a crime, at which point it fails to be “capable”.
  • You should be detecting a pattern here. We can only identify the 3 elements in the Triangle “after the event”; that is, after a crime or attempted crime has taken place – which has important and fairly obvious ramifications for crime prevention techniques and strategies (you might like to think about what some of these might be).

    2. It must be open to the possibility of falsification. That is, following Popper’s (1934) claim about the nature of a scientific theory, it should be possible for others to test it and show that it is false. It should, in other words, be open to the possibility of refutation. If a theory is not open to falsification / refutation it simply becomes something else – a statement, a claim, an argument and so forth.

    This is important because, as Sutton argues, if Routine Activities cannot be falsified and its claims cannot be refuted it cannot be considered a scientific theory with predictive value: “If an explanation for causality cannot potentially be refuted then it is not an explanation for causality, instead it is pseudoscience dressed up as causality”.

    This is something we can demonstrate using a simple Thought Experiment:

    I have a plastic gnome in my front garden, in full view of casual passers-by (I believe the public should have full and unfettered access to such cultural treasures). I leave it on show for a month with no obvious guardian to protect it.

    No-one steals it.


    Within a Routine Activities approach there are two possibilities to explain the absence of theft:

    a. There were no motivated offenders passing by my front garden in the month the gnome was on display.

    b. There were actually motivated offenders continually passing but they considered the gnome to be of no value and hence not a suitable target. Admittedly the gnome’s little fishing rod had been slightly damaged after a brief altercation with my dog, but that shouldn’t have detracted too much from it’s obvious (sentimental) value.

    The problem here for RAT is that you can plausibly suggest either as an explanation for a lack of theft. In other words, even where all three elements of the Predatory Triangle are in place the “failure to predict” a lack of theft can be plausibly, if improbably, explained away.

    Consider a second scenario:

    I decided to leave my gnome on display for a second month and within a day it was stolen.


    Obviously the “lack of capable guardian” explains the theft.

    The problem here, however, is that this illustrates precisely why routine activities is not a falsifiable theory:

  • If something happens “as predicted” the theory is confirmed: a motivated offender met a suitable target (the gnome) in the absence of a capable guardian (my dog, who I do  not keep chained-up in the garden to protect my gnome. More fool me. Obviously).
  • If something doesn’t happen “as predicted” the theory is still confirmed: in this instance, a motivated offender didn’t find a suitable target in the absence of a capable guardian.
  • If you’re not convinced by this, try a third scenario:

    Thanks to some damn fine police work, my gnome was found in a skip by the side of the road and returned to me. Determined to display my gnome but not be a repeat victim, I installed a capable guardian (a shark-infested trench surrounding the gnome).

    This seemed to do the trick.

    My capable guardian protected the gnome from theft, thereby confirming the theory.

    Until someone managed to steal it.

    My guardian was clearly not as capable as I’d previously thought.

    On the bright side, however, it still confirms the theory…

    You should, by this stage, have come to the conclusion that however this scenario plays-out, it’s always possible to confirm the theory: RAT can only ever be true.

    2. Chickens and Eggs

    Chickens. Roosting. Maybe with Eggs.

    If we accept Sutton’s argument that Routine Activities is not a theory of crime / crime prevention because it is not open to falsification, it’s still possible to see it as a model: a descriptive representation of “reality” that can be used to inform how and why we respond to “the threat of crime”.

    While this is a much weaker methodological form than a theory – models have no predictive power, for example – it can still be useful as a way of thinking about crime and areas like situational crime prevention. Even on these terms, however, Routine Activities runs into a critical problem:

    A major weakness (or strength if you’re approaching it from the opposite direction…) of Right Realist (from Broken Windows, through Rational Choice Theory to Routine Activities) approaches to crime is that they focus on the idea of “opportunity”; in basic terms, if you deny “potential offenders” the opportunity to commit a crime – through target hardening, ordered spaces, or whatever – they will not offend.

    The problem here is that even if we accept the limited parameters of the Routine Activities model the argument is a circular (“chicken and egg”) one: what comes first, the motivated offender chicken, or the suitable undefended target egg?

  • If a crime opportunity involves a suitable target and an absent guardian, an offender is motivated to commit the crime.
  • If the target is suitable, but there’s a capable guardian the offender is “demotivated” and they, ideally, don’t commit the crime (or, less ideally, they do commit the crime but their criminality is foiled by the capable guardian – which you will, of course note, still means that a crime has been committed. It is simply a “failed crime” – another potential problem that is not addressed by Routine Activities).
  • Without a motivated offender a crime doesn’t occur, but the crime only occurs because an offender is motivated by the (undefended) target -something that leaves us with a strange situation:

  • An offender only exists because there is a target.
  • A target only exists because there is an offender.
  • Each only exists because they are defined by the existence of the other.
  • Crime, in other words, causes itself…
  • While there are other problems with Routine Activities specifically and Right Realist explanations generally, particularly in relation to the various ways they theorise the role and characteristics of offenders, this post is already long (and complicated enough) so I may leave these ideas to a possible Part 2.

    On the other hand, I may not.

    Depends on whether or not I can be bothered have the time.


    1. RAT models the idea of both successful and unsuccessful crimes.

    2. It is not a theory of crime because it cannot be falsified. Whatever the outcome of actual behaviour in the real world, RAT “predicts” it.

    3. Felson and Cohen specify a routine activity approach to “direct-contact predatory crime”. It is not designed to – indeed cannot in its own terms – explain various forms of “consensus crimes”, such as drug dealing where there are only motivated offenders (buyers and sellers).

    4. The focus on an “opportunity to commit a crime” effectively means that crime comes to be seen as the cause of crime.


    Cohen and Felson (1979) “Social Change and Crime Rate Trends: A Routine Activity Approach”.

    Felson (1998) “Crime and Everyday Life”.

    Sutton (2018) “Routine Activity Theory’s ‘Mindless’ Chemistry Meme masquerades as a theory of crime causation”.

    Sutton (2012) ‘On Opportunity and Crime‘.

    Clarke and Eck (2003) “Visualising Routine Activities Theory“.

    Robinson (1998) “Burglary Revictimization: The time period of heightened risk”

    Popper (1934) “The Logic of Scientific Discovery”.

    Stay Updated

    Enter your email to be notified when we post something new:

    Archived Posts