Safety is NOT a Verb… but it should be


Subscribe to our newsletter

It may seem ironic that a guy (me) who speaks and writes for a living hated English classes as a lad.  My English grades would attest that I didn’t like English and it didn’t like me.  What I remember disliking the most was the task of outlining a sentence structure.  We had to take a sentence and place arrows to note the nouns and verbs and participles and adjectives and conjunctions and adverbs and other autocracies of school teachers trying to insert cognitive worms whose purpose was to root out a young kid’s freedom to split infinitives and worse, end a sentence with a pronoun!

I guess at one point I found words and sentences to be useful enough to begin to try to use them to communicate what was inside my head.  After all, what good is it being opinionated if you keep it to yourself?  And, to be honest, it took an old English major named Scott Geller to pound my sloppy, run-on sentences into a semblance of proper shape. 

With more experience traveling around in the real world seeing safety programs in action (or inaction) I realized that words matter.  They not only communicate but they can shape the very approach you take to your safety programming.  They can get you stuck or they can liberate your safety culture.

Consider the term “Safety” which is a chameleon of a word.  The word used in so many different ways.

“Safety” is most often a NOUN when we decree Safety first”.  This may seem like a great slogan that would inspire the workforce to think through the safety implications of their actions.  However, the great slogan may also just be a feel-good sign with no real benefit.  

W. Edwards Deming, the late influential quality guru, called these “exhortations”.  Exhortations Deming told us, give us the illusion that these outcomes are achievable and if employees simply tried harder, they would do better. This offends the worker - it does not inspire the worker.  Dr. Deming is quoted as saying:  "You can beat horses, they run faster for a while. Goals are like hay somebody ties in front of a horse’s snout. The horse is smart enough to discover no matter whether he canters or gallops, trots or walks or stands still, he can't catch up with the hay. Might as well stand still. Why argue about it? It will not happen except by change of the system. That's management's job, not the people's." As time passes the messages become washed out. Without real change no worker seriously pays attention.

“Safety” can also be a PROPER NOUN which is used to denote a particular person, place, or thing:  “Let’s call in Safety to take care of this.”  As a safety professional you should hate this use of the word because it creates the assumption that safety is a role that is done by one person or department (see Safety is Not your Job).  It’s too easy for individuals, work teams, supervisors, professionals, managers, and leaders to see safety as someone else’s job. They will wait for “safety” to come along to inspect, train, and authorize work.  This is not the type of proactive safety culture you are trying to build. Performing your job safely, making decisions that impact safety, and looking out for the safety of others is everyone’s job. 

“Safety” can be an ADJECTIVE that is used to describe a particular quality of another word.  Adjectives label.  Consider the sentence: “You are an unsafe employee”.  First of all, how can someone be un-something?  A un-person is dead.  Secondly, when you use adjectives you are labeling the subject of your sentence: “You are unsafe.”  We may as well say, “You are stupid.  Well friends, you can’t fix stupid (see Don’t Turn out the Lights by Playing the Blame Game).  When you use a label your under the illusion that you’ve arrived at a root cause of a problem.  But all you’ve done is exonerate yourself of the responsibility of finding the real risk and change real behaviors.  Your safety programs languish and safety culture becomes driven by labels.

“Safety” can be an ADVERB where it modifies verbs by indicating a place, time, circumstance, degree, cause, or manner such as in “I’m going to have to write you up for not climbing that ladder safely.”  Here safety is an outcome; Safety is the lack of injury. This use of the word “Safety” drives our measures and motivations to be outcome-based.  Traditional outcome-based measures are a rate of injuries over labor hours, severity indexes, or other rates reported upwards and outwards.  These lagging indicators are a mixed bag, its good to have a KPI that can be related to ROI to capture the CEO’s attention (FYI).  But lagging indicators do not show you where risks are being taken, only where they had been taken.  You can’t manage safety through lagging indicators… if you do you’ll be laying awake at night waiting for that phone call.


“OK guys, let’s be safe in everything we do today”.  The use of “safe” in this sentence is a SUBJECT COMPLEMENT ADJECTIVE.  It may sound good but be safe is not a call to action; it's a call for inaction.  Think about it, the best way to be safe is not to act at all, not to come into contact with hazards, and not work.  But in the work world we act.  And our actions are badly needed to create a safe outcome.  We need to engage the guards, wear PPE, read instructions, talk to others; we need to act.

In none of these grammatical uses is our word “Safety” actually doing anything.  For action we need it to be a verb.

“Safety” is not a verb. 

Instead, safety depends on behaviors and behaviors contain real action verbs.  Action verbs make them operational; when someone operates they are doing something.  That’s why in behavioral science we call behaviors “Operants.”

So consider the following sentence structure when instructing someone how to operate:

            • Do What?                 (Action Verb)

            • To What?                 (Subject)

            • When?                     (Context)

            • To Achieve What?   (Purpose)

For example:

Lock out and tag …

            the equipment energy source …

                        after your task briefing …

                   to remove the risk of energy being turned on while workers                                                     .                                                   are engaging the equipment.”


This sentence has all the components.  It gives you a clear operation.  It tells you the context where the operation should be done. And it suggests the consequence of the action.  In behavioral science we call this a Discriminant Stimulus because it helps the operator discriminate the course of action.  When presented correctly, your safety directions can be discriminant stimuli that exert control over behavior in predictable ways.   Otherwise, they can be ineffective exhortations.

So build a disciplined approach to using words to create action.  Use this sentence structure when you train, write instructions, give prompts, provide feedback and when you record behaviors in incident reports, JSA’s, and in BBS trend graphs. 

Don’t take short cuts… communicate the action.


Timothy Ludwig’s website is where you can read more safety culture stories and contribute your own.  Dr. Ludwig serves as a commissioner for Behavioral Safety Accreditation at the non-profit Cambridge Center for Behavioral Studies (CCBS: and teaches behavioral psychology at Appalachian State University, in Boone, NC.  If you want Tim to share his stories at your next safety event you can contact him at



Hi Tim,
Good point, clear communication is vital.
May I offer a few ideas, not necessarily new or mine.
1.Safety is a necessary investment not a cost. Labels impact on decisions.
2.Some appear to think constant re-labelling is development. I believe it is active non-standardisation and causes a reduction in standards.
3.Ensuring safety is the assigned task of all managers, isn’t it?
4.Ask if in your company/corporation a safety representative attends board level meetings? If not do they really put employee safety first?
FYI In aviation the ‘Definitions’ page is essential reading in any manual/publication.

Hi Tim - I applaud your raising this as an issue. Language frames thought, thought frames action. I'm not sure I agree with some aspects of your analysis, however I totally agree that this is an important discussion.

"Safety" is used as a noun ("in the interests of safety"), a proper noun as you say ("let's call in Safety") or as a noun adjunct (a noun that describes another noun e.g. "the patient safety movement"). But it is not an adjective (the related word is "safe") or an adverb ("safely").

You make an interesting point that there is no cognate verb directly stemming from safety or safe for "making safe" or "being safe". The closest would be "to safeguard" which has connotations of avoidance of harm and/or protection from attack. Certainly this is part of the familiar traditional view of "safe" and "safety" - however Hollnagel and others are taking the "safety" concept further, in wishing to encompass reliability concepts as well ("when things go right as well when things don't go wrong").

By Stavros

This is an interesting piece, mainly because the author (from fails to implement the approach he is advocating, which is encapsulated in the closing tagline:  “Don’t take short cuts… communicate the action.”  Here the author’s intended “action” is to motivate/inform people to achieve a particular (safety) outcome.  He implies that this is simply a matter of choosing the right words, without looking at the other elements of the action.  The examples cited show that it is not so simple.

In reality, the action consists of (at least):

·         Define what the objective is (what outcome is desired)

·         Determine what needs to happen for the objective to be met

·         Determine what will motivate someone to take the right actions to make those things happen (taking into account not just the desired outcome, but the complexities of human behavior)

·         Identify the best means of communicating the message that will create this motivation

·         Design and deliver the message

The tagline “words matter” applies only to the last of these elements (or perhaps the last two).  If the others are not handled properly, then the choice of words will not matter – a poor outcome will still result.

Using the example described, if it is decided that a safety outcome can be achieved by pressuring, denigrating or humiliating someone, then this can be accomplished in a variety of ways (for example, speech, facial expressions or gestures).  If, instead, the approach of describing and providing instruction on an action is chosen, it might be possible to implement this approach using gestures, pictures, video, speech or physical demonstration.  Choosing which approach to use doesn’t have anything to do with selecting words; it is a more fundamental (and non-verbal) step in the process.

Therefore, in my opinion the author has taken a “shortcut” by implying that one can achieve desired outcomes simply by choosing the right words.  This is akin to saying “cranes matter” to describe how to construct large buildings.  Yes, cranes matter, but they are not the only important element, and, in fact, some quite large edifices have been built without the use of cranes.  The right tools will fail if they are not used to implement a plan that is suited to the objective.

In any kind of communication, if the idea to be communicated is not clear and organized (or is just plain wrong), the most well-constructed message will not yield the desired effect.  I review and edit many technical documents for colleagues, and find that some can be easily touched up despite grammatical errors, awkward phrasing and other problems, because the underlying idea is being presented in a straightforward and organized way.  Other documents, with better-constructed sentences, may be much more difficult to deal with because of poor logic or organization.  In my own work, if I find myself struggling to find the right words, it is usually because I really have not thought through clearly what I wish to say, and the solution is to go back and refine the idea, not to simply seek a better mechanism for communicating the idea as it exists.

Indeed, words matter, but they matter as one part of a complex system – they are not the system itself.




I fully agree with Roger's assessment although I didn't intend the shortcut he describes.  Roger is correct in pointing out that we too often just focus on the antecedent (i.e., the "message") and short cut the motivational components of the change effort.

By suggesting the verbal components in the blog (i.e., Action Verb, Subject, Context, Purpose) I was hoping to create the context for users to analyze the three-term contingency (i.e, Antecedent, Behavior, Consequence) which are the motivational components of a behavioral system.  The consideration of that full system should suggest intervention past the simple "message delivery".

Research shows that consequence-based interventions are the strongest (e.g., performance feedback, checklists) and they are made even stronger when paired with adequate Task Clarification containing the components of the verbal messaging I describe. 

What Roger offers in addition is a link to the greater performance system by asking us to describe the objective desired and how the behaviors/actions link to that outcome.  Indeed there is an emerging set of research describing how we analyze the Behavioral Systems to include how the behaviors impact performance in ways that are important to the task, process, and ultimately, the organization.

I applaud Roger (and you) for recognizing this.  You've given me fodder for a future blog.



Thank you for forwarding Roger’s very thoughtful comments on Tim’s blog.   Roger is very astute to describe that words are part of a larger system (of contingencies) that function to alter the behavior of the listener (reader) in some actual context.   Much research shows that how we describe an event (like describing work routines in SWI’s)  can affect the way people respond to those descriptions.  A functional account of language supports his interest in designing effective communication of genuine solutions to readers who might be quite different from the author(s) of those communications. 

He is right, of course, that the solution being communicated has to be right (functional) for the situation.  To convey that optimal solution to others (e.g., the audience for standard work instructions) entails consideration of the complexities of each individual’s motivation. Roger is spot-on to highlight that.  Once the correct solution is determined, crafting messages (words, images, illustrations, etc.) to frame the communication to the intended audience is quite challenging.  There are analytic approaches to help make sense of that.

Attached are a two interesting papers that describe research into this.  These are a bit technical but convey some of the complexity of a functional account of language. Please pass these on to Roger and his team if you see fit. 

One paper (metaphors_reasoning) shows that altering just one word in a description of a problem alters the solution generated by readers.  I find it very interesting that changing one word can reframe a problem to lead to dramatically different approaches to a solution.

The second paper (RFT_IO psychology) describes rule-governed behavior and a framework for analyzing it. The meat of this paper starts on page 62 – see their analysis of why people might follow rules (or not).

Behavior science offers some very useful methodologies to inform how we frame communications with consideration of the complexity of human’s response to words.  The true test of the effectiveness of communications (verbal or non-verbal) is to observe their effect on the listener’s behavior. 

Mark Alavosius

Contact Dr. Ludwig                                 © Timothy Ludwig 2017