Interview with Steve Andreas


Steve Andreas is an NLP Trainer and developer of new patterns with his wife Connirae. Steve has produced over twenty-five videotaped and audiotaped demonstrations of NLP methods and numerous articles. Steve and Connirae are the co-founders of NLP Comprehensive in Colorado and authors of Heart of the Mind and Change Your Mind—and Keep The Change.  Most recently, Steve wrote  Six Blind Elephants, Vol. 1 and Vol. 2.  To find out more, you can visit Steve's website at


Noneuclidean Cafe spoke with Steve Andreas by phone in February, 2007.



Noneuclidean Cafe (NC): What was the genesis of Six Blind Elephants?


Steve Andreas (SA): In 2002 I published a book called Transforming Yourself, which was about how people generalize about themselves.  What are their self-concepts.  That was interesting, and I learned a lot of things about modeling self-concepts.  A couple years after I'd written that book I came onto George Lakoff and cognitive linguistics, which I'd been relatively innocent of.  And I found out they'd figured out a lot of this stuff on their own, before I re-invented the wheel.  So I became interested in a more general application of the same idea.  How do we generalize about everything, not just ourselves.  Generalizing about ourselves is very important, because it's a feed-forward system, and self-generating.  But I got interested in a wider range of how we generalize.  And that's what Six Blind Elephants is all about.


NC: I don't know Lakoff's work on linguistics, just his book on politics—Don't Think of an Elephant.


SA: Their focus is more on the linguistics and words rather than on the internal experience that goes with them, although they do touch on that.  They get there, but from the other end.  NLP starts from the individual experience, and moves from there to the linguistic description of that experience.  So they're starting at a different end of the same phenomenon. 


NC: Drilling down on some of the terminology you use in your book, what do you mean by scope and category?


SA: Scope is what's been called sensory-based experience in a lot of NLP.  Scope is a certain amount of data coming in.  If I look out right now I see my desk.  It's rather cluttered with sculptures of various kinds and with papers, a small calculator.  That's my scope at the moment.  And also the feelings I have, or what I hear.  Scope involves the different representational systems.  Each representational system provides a certain amount of information about the world around you.  So that's what scope is. 


Category is when we take a group of scopes and lump them together.  Like the word "calculator."  Almost all our words refer to categories.  If I say "car" or "calculator" or "box" or "desk," it's a category of things.  It's not a single thing.  If I point to a box on my desk and say "this box," then it refers to a single scope.  But usually when people use words they are referring to a group, or category, of objects.  And that's a collection of scopes. 


Then there are levels of categories, and that's where logical levels come in.  I can look at my desk here and I've got a certain pen.  Now, that pen is a scope.  I can see it, I can feel it, I can knock it on the desk and hear it.  So that's a specific scope.  Then if I use the word "pen" as a category, I have this pen and lots of other pens—all the pens on my desk, around the house, all around the world.  So the first logical level is when you go from a scope to a category.  At least that's how I've described it.  Now I could put pens into a more general category called "writing devices."  And that includes pencils, keyboards on computers, anything else you can write with.  That would be a more general category, and that's at a more general logical level, a higher logical level.  And then all those writing devices can fit into the category of "man-made objects."  Man-made objects could then all belong to the category "physical objects."  The entire category of physical objects would be "made of matter," so "made of matter" would be a higher category.  Each time you go up a category, you go up a logical level.  That's how I'm describing logical levels, and it is quite different than other people have described it.  It's very operationalized.  You can easily find out if someone is operating at a different logical level just by seeing if one category can be included in the other category.  Very simple test, very unambiguous.  It's consistent with logic.  Like the old syllogism, "Socrates is a man, all men are mortal, therefore Socrates is mortal."  That's a classic conclusion, and that's how I'm describing logical levels.  It's clear and unambiguous.  The other people who talk about logical levels, like Dilts and Hall, really don't have a specific operational way to describe them.  Hall has a very metaphoric way, which he thinks is an adequate description, but it wouldn't fly in science. 


NC: One of the things I found interesting is that you've added a distinction when one talks about chunking up and chunking down—and I know you're moving away from using those terms because they haven't been used with rigor—but one can talk about chunking up in terms of category or chunking up in terms of scope.  Scope and category introduce a new distinction.


SA: Yes.  As I mentioned in my book, Dilts's description has chunking down going to parts of things, which is chunking down in scope.  So he has the example of a car, which you can break down into smaller pieces, such as engine, brakes, wheels and so forth.  That's going to a smaller scope.  But then when he chunks up he uses category.  So he talks about car going to the category "forms of transportation."  That's the kind of confusion we can get into, and it's a very crucial difference.  Typically when people go to a larger category they lose track of the sensory experience of the first-level category, what Lakoff calls a "basic level" category—something like desk or rock or cat or dog.  Those categories that children start with.  It's fairly easy to collect a group of experiences into those sorts of categories.  Then as we get older we do more the kind of understanding that a car is a form of transportation. 


The other thing about categories is that we tend to represent them by prototypes.  And a prototype is sort of an average.  If I ask you to think of a bird, you probably didn't think of an ostrich or a penguin or something like that.  You probably thought of  a blackbird or robin or sparrow or something more like what you see around where you live.  So we can use prototypes when we think about categories, and this has all sorts of possible errors in terms of thinking we know what we're talking about when we really don't. 


NC: In that chapter where you talk about the way we use prototypes when thinking, you also talk about observed scope and deduced scope.  Could you say something about that distinction?


SA: Observed scope is just the way we tend to carve up reality.  When I look at my desk here, I see a little calculator. The calculator is separate from the desk.  We tend to outline things.  If there were a cat here, the cat would be separate from the background.  So that's what we actually observe.  An example of a deduced scope is hearing a meow in the middle of the night.  Now, it could be a recording of a cat, or it could be a burglar making the sound of a meow to assure me that there's no problem.  But typically when we hear a meow, we deduce it's coming from a cat.


NC: So with the deduced scope, the flow goes from some sensory-based input, to a deduced category.  Which then in turn might bring in memories of other scopes or categories.


SA: Yes.


NC: And then there was a third type of scope, based on that category we put something in, which you discussed in that chapter.


SA: Aggregate scope.  Once I put something in a category, once I put this in the category "calculator," it is associated with the computer on the other side of my desk and with IBM mainframes and with an abacus.  Any other example of a calculator gets associated with that.   So that's another way in which we tend to mislead ourselves.  Thinking we know what we're talking about. 


And in terms of politics, the kind of words the Pentagon uses, they're deliberately used to avoid having a specific scope of experience.  They talk about "collateral damage."  Collateral damage of course means dying human beings, blood all over the place, body parts.  But when you use "collateral damage" none of that scope comes in your mind typically.  So it's a nice, safe word.  Collateral damage, nothing to get upset about.  That's one of the ways we delude ourselves. 


NC: If we look at another term from your book:  "frames."  It seems you are using "frames" in the way a traditional NLPer would understand it, but there's an added distinction, between scope and category.


SA: When I use the term "frame" I'm referring strictly to scope.  If you look in a given direction, you put a frame around what you're looking at, and you exclude whatever is outside that frame.  That's a scope limitation.  I used the word "bundle" as a way of saying a category is a bundle of stuff.  That word "calculator" creates a bundle of all the calculators I've seen or heard of.  So a category is a bundle of scopes. 


What's traditionally been called "reframing" is a mixture of both.  And then, even in categorical reframing, there are different logical levels of reframing.  So you can reframe at the same logical level, at a smaller logical level, or at a higher logical level.  It's a little like DNA.  There's only four codes, but out of those four codes you get all the multiplicity of life forms that have ever lived, and that are going to exist after us.  Same here, you start with a simple distinction.  I've found it a very useful distinction.  For me, it's way of organizing everything I do.  Not only in thinking about things, but in doing therapy, change work and so on.  What scope is someone attending to, and how are they categorizing it?  And on what logical level are they categorizing it.  That gives me three flavors.  I can change their scope, I can change how they categorize it, or I can change at what logical level they categorize it.  As far as I can tell, every piece of therapy, every piece of personal change—whether intended or spontaneous—includes one or more of those processes. 


NC: I know one of the things I enjoyed in your book, was that once the terminology was in place, you went through and showed various changes, NLP presuppositions, NLP processes like timelines, could be understood in those terms.


SA: I haven't found anything I can't understand in those terms.  They offer a handle.


NC: I know I like finding handles like that, ways to look across multiple processes.  Without some sort of handle, it's easy to lose track of what exactly you're trying to do.


SA: A lot of people out there don't know what they're doing.  They'll apply processes or work with patterns, without a clear enough understanding of whether it's what the client needs or not.  For me, this offers an extremely useful way of understanding what you're going after.  And then the processes and patterns offer a range of ways to do that.


As you mentioned, the two volumes give a number of processes and change work—family therapy, individual work—that offer examples of the scope and applicability of the concepts.  It's a little like energy and matter.  Energy and matter, as far as we know, comprise the whole universe.  Everything can be described in terms of energy or matter, or some of both.  And that kind of variable, which is widely applicable, is tremendously useful in organizing all the different things we see and hear and feel.  To realize they're all examples of energy and matter. 


So I'm having a lot of fun with this.


NC: I found the distinctions quite useful.  So thank you for writing the books.


SA: It was a lot of work, but it was also a lot of fun.


NC: Now, there was one thing you said in the book, that I thought made a nice transition to another things that's of interest to me.  And given your science background, I was curious what you thought.  You mentioned the difference between faith-healing and science.  If an NLP process works, it doesn't depend on the person's belief, just their cooperation.


SA: Right.


NC: I quite like that statement.  However, a number of the NLP teachers I've known, and the NLPers I've talked to, seem interested in pulling in the placebo effect where possible.  They believe the person's belief is important.  So I was wondering what you're take on that was.


SA: It's interesting, because the placebo effect, presumably, directs people's attention to a different scope of experience.  They think they're getting a pill for whatever medical problem they have.  So their internal experience probably is changed in some way.  They are looking at scope and category in a different way.  So probably faith healing can be described in terms of scope and category. But I was trying to make a distinction there.  In any of the branches of science, assuming it works, you have processes that work—bridges stand, mathematics actually describes a certain set of events, whatever it is.  That description is enough to make a change.  Once you describe a process, then you don't need faith healing.  You don't need to have faith in your cell phone, it just works.  As long as the battery is charged, no one has stepped on it and broken it, you can dial all around the world.  Which is really incredible.  The complexity of the math that goes into that is just mind-boggling.  I have only a small idea of that, but given that I have a degree in chemistry from Caltech, I do have a pretty good background in hard science; I read the journal Science every week. 


Science is about finding lawfulness, and finding what works.  One of my favorite quotations is, "The word 'science' should only be given to the aggregate of recipes that always work.  Everything else is literature."  (Paul Valery) And unfortunately in psychotherapy there's an awful lot of literature.  In fact there's an approach called "narrative therapy," where the whole goal is to retell someone's story.  Now, retelling someone's story could be understood in terms of scope and category, too.  If you retell someone's story, you tell different content, and different content is a different scope of experience.  You summarize it in different ways, categorize it in different ways.  So that could be useful.  But in science, you need to make a specific recipe.  Science is injunctive: Do this, and this will happen.  Do that, and that will happen.  Do this series of things, and this is the result you should expect.  And if you have that, then you really don't need faith healing. 


But presumably, that same kind of scientific analysis could be applied to faith healing.  I don't know much about faith healing.  But let's say the placebo effect.  That's been studied somewhat.  If we could go in there and find out that when a placebo works, it's because the person who received the placebo gets thus and such different image in their mind, different scope of experience, they think about it in a different way, they categorize it differently, and so on.  If we could do that, then that would turn faith healing into science.  But what people usually talk about in faith healing is laying on of hands and then everything will be fine, you'll be all right.  And that is not science.  Even if it works, even if it works let's say erratically, or for some people, it's not science. 


NC: Continuing on the subject of science, and looking at NLP.  And if I've missed a body of research, please correct the assumption behind my question, but it seems there's no scientific evidence that NLP works.


SA: I think you're right.  There's no good science behind NLP.  Twenty years ago I surveyed everything that had been done in terms of trying to validate, or invalidate, NLP.  And even the studies that showed that NLP worked were flawed, very badly flawed.  It was lousy science.  And most of them said it didn't work.  But they did all kinds of things that just were nuts.  They tested hypotheses that no one in NLP would say were reasonable hypotheses.  They didn't have good ways of testing them.  Even if an hypothesis was one that was valid in NLP, they didn't have a good way of testing it.  They used paper and pencil documents instead of noticing the cybernetic loop between the experimenter and the subject.  It was just lousy work.  


NC: As someone with a scientific background, does that concern you?  Do you think other things have proven NLP?  Or are you hoping that sort of scientific research can be done in the near future?


SA: I'm certainly hoping it can be done.  The IASH group, the Institute for the Advanced Study of Health...


NC: I was very hopeful when I saw their recent email that they would start pushing NLP in a research-based direction.


SA: Yes.  There's a man named Frank Bourke, who's working very hard to get something set up.  He doesn't have any funding at all right now, but he's got a lot of people who are connected with grant writers, and they're trying to get four or five million bucks to start doing some heavy-duty science.  And that's the main problem—it's very expensive.  Even a small pilot study, you have to hire people, secretaries, etc., to process the data. 


My own conviction, given that I've cured a whole bunch of phobias in a row, is that I'm reasonably sure that it works.  But I don't have scientific proof.  No double-blind study.  That's what's needed.  And hopefully we'll get something out of this.  I'm putting a good deal of effort into setting up the protocols, looking to validate NLP.  So we'll see. If we can get some money to run a bunch of people through, I'm sure we'll get some good results.


NC: Since I know many of our readers are doing change work, I was wondering what are the things you look for when doing change work with a client?


SA: Along the lines of what we discussed earlier, I'm looking at what scope they are attending to, how they are categorizing it, and what logical level they're categorizing it at.  Those three variables.  So for instance, if someone is grieving, they are paying attention to the scope of experience they've lost.  They're typically seeing it disassociated, far away, and no longer available to them.  Well, that's a mistake.  It's available to them.  They can go back to that experience and experience the good feelings they had with that person.  But they don't realize it.  And the grief process my wife and I worked out fifteen, almost twenty years ago, is very simple.  You re-associate, which is a change of scope, because you're inside the experience rather than at a distance.  The content you experience is different, and the main way is that you can access those wonderful feelings you had with the lost person.  When you're disassociated, you leave those good feelings behind.  You can see yourself over there in the picture having a good time, but you're not experiencing it.  So every NLP process, you can just pick it apart and say, here's where the scope changes, here's where the categorization changes, here's where the logical level changes.  It means that all the apparently different interventions, all really have a common foundation in one or more of those three variables. 


NC: One question I like to ask is:  If, knowing what you know now, you could go back and talk to yourself when you were just starting your career, what would you like to tell that younger self?


SA: Hang in there throughout the confusion, there's a lot of good stuff ahead.  And that applies to someone who is just starting now.  There's an awful lot of stuff that's already been discovered, and categorized.  And then there's even more ahead.  We're just beginning.  Physics is about two hundred years old, give or take twenty or thirty years.  And two hundred years ago, it was a lot like NLP, there were a few people diddling around in their labs, fiddling with batteries and electro-magnetism.  And they really didn't know what they were doing.  They were scattered all over Europe, and a few in America.  Two hundred years later, physics is incredible.  They're doing amazing things.  And the same is true, I think, of NLP.  What we're doing now will be looked back on in fifty or a hundred years, and they'll say, "Boy, that was crude."  Just take something like the visual squash.  The visual squash works, but it's very crude.  It slams a whole bunch of experiences together, and then the person goes into a trance to sort it out unconsciously.  It's a really crude approach, like some of the early medical techniques, where they really assaulted people's bodies.  Sometimes it worked, but it was still a brutal assault.  Medical technology has gotten much more sophisticated.  And they've still got a long way to go, as old as medicine is.  Modern medicine is at least a hundred years old, and it's got a long ways to go.  So there's a lot ahead.  We're just beginning to scratch the surface, I think.


NC: And finally, anything else you'd like to add, that my questions didn't lead to?


SA: Nothing in particular.  I'm just totally involved in this new way of looking at all the stuff we've got, and still having a lot of fun with it.


NC: I know fun is an important criteria for me in these things.


SA: Absolutely.  Some things are work and they're also fun.  But if it's not enjoyable, why do it? 


NC: Thank you very much for your time.


SA: You're very welcome.


Previous  Home  Table of Contents  Next


Copyright © 2007 James Swingle. All rights reserved.