Proving Safety (2): A three part series

June 18, 2024



This is the second in a series of three podcast with Greg Smith, the author of the book Paper Safe. In this podcast Greg and I discuss his new book Proving Safety, why he chose the title and the subtitle, and how is the book different from his previous book Paper Safe.

Further information



Nippin, Greg Smith


Nippin  00:00

Nippin Anand, welcome to another episode of embracing differences with me. Nippin Anand. Anand, a podcast aimed at understanding and promoting transdisciplinary ways of living and thinking, meaning, assimilating different viewpoints, different subjects, different disciplines, but focused on a very simple question, how do we human beings, learn, unlearn, relearn and make decisions, and how can we tackle risks in an uncertain world? This is the second in a series of three podcasts done with Greg Smith, the author of the book Paper, safe, and, more recently, proving safety. Given the amount of time that many authors spend in coming to terms with what would be the most ideal title for their book, I thought it would be worth delving into. The reason behind Greg coming up with this title, proving safety. And also, the other thing that we question here is that, how is this book any different from the book that Greg wrote earlier, which is called Paper save so I hope you enjoy listening to this podcast as much as we enjoyed creating it together, and stay with us until the final third podcast, which is where we talk about the notion of critical risks, because we have an important announcement to make. Here we go.


Nippin  01:36

Let’s start off with another question, and I think this is something which is very close to your heart, probing safety after paper safe. This is another blockbuster, I would say. But I have


Greg Smith  01:52

blockbuster in safety. Let’s just gave it all right, okay, let’s put paper rail.


Nippin  01:56

But no, I just look at the title of this book, Greg and I wonder, what was the motivation to come up with a title like this? That’s my first question. Okay, explain that to me.


Greg Smith  02:12

I people ascribe a level of conscious thought to some of what I do that doesn’t exist. So proving safety just rang well, and I was mucking around with titles. And my middle daughter, who’s a playwright, thought it was quite clever that I had paper safe PS, followed by proving safety and in alphabetical order, entirely coincidental. Entirely coincidental. If you were, if you’re going to be more precise, a better title might have been evidence in safety, rather than proving safety. But at the same time, I think if you, if you look at proving safety from a health and safety industry perspective, that’s probably not a thing we can do to prove that we’re safe. And as you know, as a chapter in the book about what is safe, but from a legal risk management point of view, you can prove that you’re meeting your legal obligations. And so in that narrow legislative context, you can prove that you’re safe. It’s just that our systems are typically not designed or set up to do that, even though, I think most organisations think that’s one of their primary purposes.


Nippin  03:30

Can you unpack that a little bit more? Because, on the one hand, you’re saying that it is possible to prove safety, yeah, except that our systems are not designed for that. So Can you unpack that, that paradox a little bit more well?


Greg Smith  03:47

Again, it’s in the book, obviously, but it’s not new. It’s a conversation I’ve been having for decades to say to both to clients in in case, in actual prosecutions, and in the context of training and advice work, but I can’t. I’ve never been able to pick up a client’s monthly health and safety report and say This provides information that I need to demonstrate legal compliance. I’ve never you won’t find a prosecution in Australia where the Defence Council refers to a client’s injury rate data or health and safety metrics as evidence of compliance with their legal obligations. It just doesn’t illustrate or evidence any of those things, even even if the the data in those metrics and indicators was not corrupted, which I think is unlikely, it’s still not evidence that the systems are safe. And yet, I I. A, I suspect, and B, I have have had conversations with senior leadership to this effect that they see one of the primary reasons that their safety management systems exist is to evidence and demonstrate legal compliance. And when you sit with I’ve done prosecutions with small businesses who have got a set of folders that they’ve bought from a consultant. And I remember one conversation in particular where they said to me, Greg, we actually thought that all we would have to do is slide these folders across the table. The regulator would take them, and that would be the end of it, which is kind of where paper safe was born from that disconnect between the documented processes and our level of compliance and so and this sort of goes back to the conversation we’re having a little while ago, this idea that, do organisations actually really want to know the state of safety in their organisation? Do they actually want to understand if it’s achieving the outcome they want it to achieve? And I’ve said a few times, both in writing and in public forums, that if organisations really wanted to understand if safety management or if they were managing the risks in their organisation if they really wanted to understand that, then health and safety reports couldn’t look like they do, and injury rate data could not still have the central role that it does in our organisations. Because if you look anybody who looks at injury rate data for a second must understand that it doesn’t give them the information they need to understand if their systems work. That’s not a that you don’t need to be a lawyer to do that, or particularly intellectual. You just need to look at it and say, What do I need to know? Well, that doesn’t tell me that. Yeah, so I think that’s part of it.


Nippin  06:56

So the reason for writing or having a title like this is primarily centred on the idea of, how can we prove safety within a legal context? Yeah,


Greg Smith  07:09

I think if we’re talking about proving safety, it’s primarily from my perspective in that legal framework. Part of the argument is to say, because for a long for the longest time and as long as I’ve been involved in it, legal compliance has been misunderstood, for starters, and also, legal compliance is seen as much lower on the hierarchy of morality. All right, so our legal obligations are our minimum requirements. Now I don’t believe that’s true, but let’s just stay with that thought for a moment. If you are a health and safety manager who says legal compliance is our minimum requirement, all right, then surely you can demonstrate it. If that’s the minimum you have to do, then you should be able to demonstrate legal compliance, and we know that most organisations cannot. So I think calling it your minimum requirement is a bit of a cop out at one level, but then the other level is and this is the dismissiveness of legal compliance. So and a bit more. There’s some of that improving safety, more of it in paper, safe. But this idea of a checklist, okay, you complete a checklist, there and goes, I tick, tick, tick, tick, tick. They’re just doing that for compliance. They’re not if they’re just ticking it. That is not compliance. Compliance is understanding if it works. And so I think the health and safety industry kind of steps its way around the difficulty of legal compliance by saying that it is something mechanistic and simplistic and not real, and that’s that’s not what it is. I mean, the test under Australian legislation and several jurisdictions is reasonably practicable, if you dissect that as a concept, what you are talking about is, does the organisation have proper systems to manage the hazards, and are those systems in place and effective? And I think that any organisation that could demonstrate those two things would have a safe workplace. Now what happens then is that people equate proper systems with highly documented processes and proper supervision or adequate supervision with sort of mechanistic checklists. And that’s not true, that there’s no mandated way that you have to do this. So if you take the new view of safety that. Says we need we need adaptability. I can’t remember all the terminology. We need resilience, but one of the things they talk about is an ability to vary locally. Do we don’t want prescriptive systems. We want our teams to be able to vary locally to manage the risk. That’s fine if you can demonstrate that your work groups, varying locally, are meeting the requirements of the regulations and codes of practice, and that that is effective to manage the risk. That’s fine, that’s reasonably practicable. That’s an acceptable way to manage the risk. But ticking boxes on forms and completing paperwork without understanding cause and effect and effectiveness is not legal compliance. So I think there’s a misnomer there.


Nippin  10:51

Yes, I get it. Maybe let’s just talk a little bit about what you just touched upon, which is local variability or whatever the term is. In essence, what we are talking about is that improvisation, yeah, at a very, at a very local level, yet at the level of the of the person who’s doing the work


Greg Smith  11:12

that happens, that happens naturally, yeah, yeah. And that is


Nippin  11:15

something which, which is, which helps manage safety or, yeah, and what you’re so. So what is the legal perspective on this? Maybe we can unpack that little bit here, because I think that’s a really important part


Greg Smith  11:33

to understand. I don’t think the court particularly has a view about variability being right or wrong. So the court and again, variability is an interesting thing. So if you take something like working at heights, you’ve got a code of practice that says if you’re working at heights, you need fall arrest or fall restraint. That’s what the code of practice says. So when we talk about varying locally, we’re not saying there is a discretion to decide locally whether you use fall arrest or fall protection. That’s a basic given. The court’s not going to give you much leeway in that context. That’s expected to be seen. I remember having a call with somebody talking about excavations greater than 1.5 that they weren’t barricading, and this was an organisation quite heavily invested, I think, in some of the hop principles. And it was like the conversation was, we know it’s in breach of the regulations, but here are all the reasons why we understand it occurs. I said, No, no, none of that’s a defence if the regulation says over 1.5 metres you have to have these controls in place, then you have to have those controls in place, all right, but just having the controls ostensibly in place is not compliance. You still have to know that they’re in they’re effective to manage the risk, but no local variability or workers adapting to the situations they find themselves in is perfectly acceptable. The difficulty I think we create for ourselves as organisations is we write procedures with no discretion built into them. So I think we can, we can accommodate variability, so long as we are very clear about what are the not negotiables in a work practice, and what are the elements where we’re entitling supervisors to exercise their discretion. Now, I came very much face to face with this about four years ago. Now, I was dealing with a I didn’t I did the investigation. I wasn’t acting as the lawyer. Another law firm brought me in to do the investigation. I was a pretty horrific accident, and on any view of the evidence, whether there were contributing organisational factors or not, the supervisor in charge of that job who was killed had evidence significant non compliance with the organisational processes, Regardless of how effective they were significant non compliance, and had positioned themselves in a way that when they when they gave the direction for the work process to recommence, the failing was always going to result in their death. So it was problematic. And I remember presenting findings to the board, including the chief executive officer, and talking about how we needed and I was probably a little bit caught up in some of the new ideas at the time, and talking about giving supervisors more discretion, and the CEO, who’s very. Well regarded CEO. She stopped me sort of mid sentence and said, But Greg, isn’t that exactly what happened here? The supervisor exercised their discretion, and she was right in that sense. Now you can talk about organisational factors that might have influenced the discretion and all of those sorts of things, but if we are going, if we are going to say that part of our health and safety strategy is to give more discretion to the front line to make decisions about how they do work, which is what they are doing anyway, all right, then we better make bloody sure that we’re equipping them to make those decisions. And I think that’s probably part of the part of the risk here.


Nippin  15:45

And I think you’re absolutely right. I agree with you in more than two counts. One is that we need to have some sort of a mutual understanding about, a shared understanding about what that discretion means.


Greg Smith  15:58

What are the boundaries of the discretion? Yeah, I think that’s right.


Nippin  16:01

But I think the other important thing is, and I think this is where we touched upon it during lunch, is that, how do the processes, the documented processes, actually manifest that? How do they understand? Yeah,


Greg Smith  16:17

but, but again, whatever approach we take the onus is on the organisation to equip the people for success. So it’s very easy to be critical of things like Jha and safe work method statements and documented processes because they’re not fit for purpose for the workers who have to implement them. I think that’s quite true, but equally and as problematically, if we say we are going to give discretion to the workers at whatever level if we haven’t given them the training and competence and understanding and experience to exercise that discretion, well we’re going to have the same problem, because ultimately, and and if the organisational influences are what we’re led to believe they are, then whether we’ve got documented processes or the exercise of discretion, if the behaviour is being driven organizationally, it doesn’t matter how we do it. Such


Nippin  17:15

a good point, Greg, because in a world, I mean, if you look at the two the last two or three aviation accidents that have occurred in high profile accidents, one thing that comes out is the aviation industry struggling with the competence of people, experienced people, competent people. And on top of that, if you put a layer of something like what you just described, which is local variability or adaptation. I think this falls flat on its face, because you are stretching. You’re already stretching the boundaries of your operation by putting people who probably don’t have that level of experience and competence, although they might have a certificate to operate, but they may not be, no, they may not be the kind of people who are actually comfortable in with that kind of variability. So you are sugar coating something which is inherently dangerous,


Greg Smith  18:11

yeah, potentially and again.


Nippin  18:16

And so my point really is that I think one of the reasons why this appeals so much this this concept appeals so much to businesses, is that they can actually play around with these ideas and make it work for the business. Yeah,


Greg Smith  18:32

yep, I don’t dare then I think


Nippin  18:34

that. So my big problem is that, when things go wrong, how much of this really helps? I mean, in an accident, for example, well,


Greg Smith  18:43

yes. So again, the the courts are reasonably agnostic in terms of how you do it, but if you are going to stand up in front of a court with a safety philosophy that grants a lot of discretion to the workers to decide how to perform the work, then you still need to demonstrate that that was a proper system to manage the risks, and it was implemented and it was effective, and again. But there’s the language of this conversation gets really confusing. So it’s one thing to say, we’re going to consult with the workers to come up with the better processes, but again, you still have to have regard to the legal requirements, okay, so you can come up with the processes and decide the process, but then you still have to know that that process is implemented and fit for purpose, and then The other so that that kind of some middle ground, and then at the other end, you’ve got process developed on high and imposed, which has got problems. And then what seems to be emerging at the other end of the spectrum is local variability. Workers decide how they’re going to do the work. Work, which I think has equal numbers of problems, and those things all exist on a spectrum, the spectrum. But ultimately, what is, in my experience, what is missing from the conversation is, what is the assurance mechanic to know that it’s achieving the outcomes we want it to achieve. And as I said, I’m not convinced we do that at all well under any methodology, no.


Nippin  20:28

And it’s really, really important point from from a legal perspective, because what you’re suggesting is that all the contemporary discussion that we are seeing around risk and safety, or health and safety, the it is so distanced from the legal framework in terms of how it should work in practice?


Greg Smith  20:49

Yeah, yes, I think there seems to be a disconnect


Nippin  20:53

here between what the innovation where the innovation is leading, and what the regulatory framework is really geared to,


Greg Smith  21:02

oh no,


Nippin  21:04

not so much regulatory, but


Greg Smith  21:05

the legal framework. I think, I think that’s right. I think that I am reasonably I’m reasonably comfortable to say that, particularly in an Australian context, the development of health and safety legislation over the last decade is not consistent with contemporary safety science. I’m pretty comfortable with that as an idea, but at the same time, I am yet to be convinced that the health and safety industry as a matter of practical application of their ideas has given enough emphasis to understanding whether these things achieve the outcome they’re designed to achieve in Yeah, when they’re when they’re when they’re taken out of a and it’s not laboratory, is not The right word, but when they’re taken out of a research setting, how do we Tran how do we put those into an organisation that doesn’t have academics overseeing the exercise? To understand is that achieving the outcome and maintaining the achievement of an outcome.


Nippin  22:19

So I see at least two problems here. Not only that the legal framework is not geared up to address some of those risks as a result of safety innovation. It’s also the fact that a lot of innovation, perhaps, is disconnected, or it’s it there are some gaps between innovation and how it is implemented in practice. Oh, you’re saying, I


Greg Smith  22:39

think, I think that second part is, is true, yeah, and I think that’s always been the case. And I think the challenge with safety innovation is that it tends to devolve over time. So if you take any safety practice historically that’s been introduced into an organisation. I think the trajectory is kind of consistent, and it devolves to a greater or lesser degree and more quickly and more slowly, into very much just a process driven exercise, as opposed to a outcome driven exercise. Can’t


Nippin  23:19

agree more, Greg, because that’s precisely why where my energy is focused is at the end of the day, it’s about having those conversations that bring the two together, and the ability to actually demonstrate that in practice when shit happens. Yeah,


Greg Smith  23:36

I think, I think that’s important, but, but even, even then there’s like, there’s another step to go, I think, because I think we get lots of very good individual practitioners who can go and have conversations and get a good sense of what’s going on in that conversation. But the institutional challenge is, how do we aggregate up that collective knowledge to create an overall understanding?


Nippin  24:06

Yeah, absolutely. And maybe that’s a top one of the discussion, because I had certain things in mind to discuss with you, but we talk about that, but bringing back to proving safety, yeah, this is where we started off from. So how does this book actually address some of those issues that we just discussed about.


Greg Smith  24:25

I guess the book does a couple of things. One is, it tries to set out the problem of, what is safety? So what are we talking about? It tries to put it into the context of wicked problems and the challenges that come out of that. It then has a very specific focus on an Australian legal context to talk about reasonably practicable and due diligence and challenges the current state of metrics around those but ultimately nippin What it proposes is a framework with a few suggestions in there about doing less but doing it better for. Doing things with intent and purpose. I think the fundamental message of the book is to move assurance activities away from a metric collection exercise to the creation of narratives describing what we see. And I hope, I hope, one thing I would like to see picked up from this piece of work, is a suggestion that I’ve put in there that if whenever we’re inquiring into work, so whether it is after an accident, and it’s whatever type of inquiry you choose, or even when there’s no accident, we’re just inquiring and understanding work, to have two issues for foremost in mind, and one is and one is to understand, is this a fit and proper purpose to manage this work safely? And prior to this inquiry, what evidence do we have that that system was implemented and effective? And I think if we can just start to integrate those two questions into our safety thinking, that will go a long way to demonstrating efficacy.


Nippin  26:11

Great. And I think that’s a nice way to end it. The book is here, and if you wanted to read it, yeah, it’s a wonderful opportunity. I’m going to read it on my way back home. Got


Greg Smith  26:23

your flight home? Yeah?


Nippin  26:25

Gravity, yes.


Greg Smith  26:26

Thank you. Plenty of time to get through it. Thanks. When are we going to hold yours up?


Nippin  26:36

If you’ve enjoyed listening to this podcast, many more podcasts are available on our website, novellas, dot solutions, forward stroke knowledge space, the podcast embracing differences is available on Spotify, podbean, Apple podcasts and anchor. You can also subscribe to our YouTube channel, Team novellas, that way, every time we publish a new podcast, you will get to know you want to find out more about our work, visit us at novellus.solutions, or simply write to us at support@novellus.solutions. Thank you for wanting to learn more than you knew yesterday and until we meet again. Goodbye and have fun. You.