# Map and Territory redux

**URL:** <https://boards.straightdope.com/t/map-and-territory-redux/129325>\
**Category:** Great Debates\
**Created:** [September 23, 2002, 7:13pm UTC](https://boards.straightdope.com/t/map-and-territory-redux/129325 "2002-09-23T19:13:06Z")\
**Posts on this page:** 20\
**Page:** 1

<div class="post-metadata">

**Author:** ![erislover](https://avatars.discourse-cdn.com/v4/letter/e/71e660/32.png) [@erislover](https://boards.straightdope.com/u/erislover)\
**Post date:** [September 23, 2002, 7:13pm UTC](https://boards.straightdope.com/t/map-and-territory-redux/129325/1 "2002-09-23T19:13:06Z")

</div>

This thread was lost, so here I am again.

Now, we have an epistemology, a theory of knowledge, and from this epistemology we say we know something. That is to say, the proposition P is true.

What is the status of the proposition P? And I mean this in several ways, though the ontological sense is what I am mostly aiming at here.

Consider that our knowledge—that is, the “set” (I use the term loosely) of propositions which have been assigned a value of “true” by our epistemology—is, as it were, a _map_ of reality. And, quite literally, as we would point to a map of San Fransisco and say, “This street is here”, so, too, would we access our epistemological statements and say, “This [proposition] is true.”

And “true” here means, of course, corresponds to the real world (however it exists which would probably also be known epistemologically). And it is just this correspondence that gets to me.

How does a map _correspond_ to a city? How does a proposition _correspond_ to reality? I don’t necessarily intend this as a question about meaning. If we conceive of a (again, loosely described) knowledge-set as _a map_, then I think this question needs to be answered.

For what is important to me here is this: knowledge of objective reality (or even inter-subjective reality, as the case may be) is just that: objective. If I say “I know P” of course the issue here is P itself, and not even that. The issue isn’t _the proposition_, which we should like to say is only incidental, but rather the “referent” or “target” of the proposition. In the analogy of “map / territory”, we talk with and point at the map, but of course the map isn’t our concern. “This street is here” isn’t meant to say, “You will find a mark on this paper” but rather “In the city of San Fransisco if you enter in such a manner you will find a street there.”

Normally this doesn’t concern us. But some like to stress the notion that the map is not the territory (from the creator of General Semantics… thanks **Duck Duck Goose** , you’re a princess!). And now we start to run into problems, IMO.

Reality = territory. Knowledge-set = map of territory. Epistemology = map of the map of the territory. Proposition is a map of the map of the map? And if I say “P” and you hear “P” now do you hear the same thing I say, or is this a map of a map of … [etc].

I do not appreciate such limitless recursion, for one thing, and for the other this means that every “translation” that goes on further removes “what we know” from what we can say is the referent of what we know. This seems to paradoxically lead us to say that the more formal an epistemology is the less it says at all (which, now that I put it like that, spawns some other thoughts about the notion that the propositions of logic say nothing, but nevermind).

So, let’s back up here and look at where we stand. We have a metaphycial existent: me. We have another: that street. “I know that street is there,” sez I, and here we see that it doesn’t really matter whether I in fact am holding a literal map or not. The question is: what status does this statement have? If we take the map / territory tact we find maps _everywhere_, including the analysis of “what it is to correspond to something”. In other words, it gives us infinte recursion with “corresponding” being completely undefined or _a priori_ (and I am not very keen on the _a priori_ so I’d prefer to leave that option out).

Of course, we can also think of propositions like this. “I know p.” Now, can I also say, “I know that ‘I know P’” is true? Hofstader in Godel Escher Bach thought about such statements. If we extend the set-theoretical analogy we would sort of find ourselves discussing (perhaps) a sort of theory of types, and propositions about propositions are an order higher than propositions themselves. Which, I think, is one tolerable (though misleading IMO) way to tackle the issue, but…

## …But this requires a sort of atomic proposition, now, doesn’t it? And this is the very thing.

Consider: a dictionary. A dictionary consists of words defined by words; however, most of the words we learn that are necessary to utilize a dictionary are, of course, not learned tautologically (which a dictionary essentially is) but rather by use and pointing. That is, the self referentiality (the infinte recursion of looking up a word, then looking up the words that are used in the definition, then looking up the words for that, and so on—a process that simply will not end though may become cyclical after a point) is only so when we think of the dictionary as defining language. Of course it doesn’t; we can only use a dictionary after we know a certain portion (which isn’t to mean a specific or explicitly definable vocabulary) of the language already. Language doesn’t “bottom out” on its own, but it bottoms out in experience of the real world (again, whatever that is).

So a dictionary is a map. But to view it like that is to miss the very point of “corresponding”, doesn’t it? Because a dictionary, like a map, is self-contained. It doesn’t _have_ to attach to reality or correspond to reality in any way (consider maps in fantasy books).

“The map is not the territory.” Fine. So what is it? “I know P” isn’t used to say something about me, it is used to say something about P. So here is the proposition G: “I know P.” _What_ is G? Does it have a sort of transcendent existence? Is it an immanent property of the speaker’s (in this case: me) consciousness? A map is a map _of_ something… but what is this “of” consist in? The map _attaches_ to reality in some sense. In what sense? If it _doesn’t_, then of course the map is not the territory: _there is no territory_. If it does, then this “correspondence” presents us with a sort of ontological problem.

At least, it presents _me_ with one. What do we count as a representation? —Our sensations (the phenomena)? What do they represent—explanations end somewhere—and how do we know they do if our knowledge itself is only a representation? Somewhere we must have “the thing”, the bottoming out, if we look at knowledge as recursive.

---

<div class="post-metadata">

**Author:** ![erislover](https://avatars.discourse-cdn.com/v4/letter/e/71e660/32.png) [@erislover](https://boards.straightdope.com/u/erislover)\
**Post date:** [September 24, 2002, 4:10pm UTC](https://boards.straightdope.com/t/map-and-territory-redux/129325/2 "2002-09-24T16:10:58Z")

</div>

Perhaps for clarity I will state this.

If knowledge is a representation of reality (but it isn’t reality) then how does it attach to it (which is to ask, how is it a _representation_ if we don’t have access to things-in-themselves)? For isn’t an understanding of the attachment then only another representation (isn’t understanding of ‘understanding’ only a map and not the thing itself)?

But to resolve this, there must be some sense in which the representation itself exists, a sort of pure intermediary, some sort of atomic propositions which not only _represent_ something but can be used as the thing (for example, that I would treat my knowledge of a bed the same way I would treat a bed; of course in this example we see the distinction clearly).

Contrariwise, we could accept transcendence of our consciousness (Husserl’s transcendent ego, for example), of our propositions, of reality (and go headlong into platonism). We could also simply, like Berkeley (and Hegel _et al_) resign ourselves to idealism; we could, as it were, banish the idea of _matter_ and accept only _mind_. Here there is no territory, only the map.

None of this feels satisfactory to me.

---

<div class="post-metadata">

**Author:** ![ElJeffe](https://avatars.discourse-cdn.com/v4/letter/e/958977/32.png) [@ElJeffe](https://boards.straightdope.com/u/ElJeffe)\
**Post date:** [September 24, 2002, 8:20pm UTC](https://boards.straightdope.com/t/map-and-territory-redux/129325/3 "2002-09-24T20:20:59Z")

</div>

Wow. Deep. Okay, here’s a few random thoughts which (I hope) at least begin to address what you were talking about:

As far as what a map is, let’s suppose an infinitely accurate map of San Francisco. (Since San Francisco is constantly changing, we can imagine that this is an accurate map of San Fran, as it existed at some point in time.) Everything you can think of is represented by this map. The position of every street, every building, every person, every dust molecule, is represented here. Technically, this would require a 3-D map, or an infinite number of 2-D maps, but this is just a thought experiment, so you get the idea. This map then presents us with a one-to-one correspondence between reality and paper. Point to a spot on the map, and that represents, say, a point on a street. Every point on the map can be… well, mapped… to a point in San Fran, in the same way that every integer can be mapped to a phrase that describes that number (ie, 1 → “one”, 56 → “fifty-six”, etc.)

Now let’s move to something a little more realistic. Suppose that someone goes and constructs a super-accurate top-down map of all the streets in San Francisco, confined with limits in current technology. This map is no longer going to be perfect, but it’ll still be accurate. The width of any street (as measured by the scale of the map) will be close to reality, but always a little off. How does this compare with the Perfect Map? You can still construct a one-to-one correspondence between elements, but you can no longer do so to the same degree. Every point on the map no longer represents every point in the city. However, you can point to, say, a tiny area on the map, and map that to a tiny area of San Fran. Say, a one-foot-by-one-foot area in the city, versus the corresponding area on the map. You can no longer, however, provide an infinite number of such mappings. There’s a smallest subdivision you can get to and still have a true one-to-one correspondence between the map and reality. If, for example, the map is drawn to be accurate to within 1", then you can only provide one-to-one correspondences for a finite number of 1" squares, while still maintaining “truthfulness”. If you try to create one-to-one correspondences on a scale smaller than that, you sacrifice accuracy. Some of your correspondences may reflect reality, some may not. Still, you can make an attempt to map on a smaller scale than 1", and if you stay pretty close to that scale, you’ll probably be more-or-less accurate. If you try to map 0.75" squares, for example, you’ll probably be pretty accurate overall, and in most cases where you’re off, you won’t be off by much. A few of your mappings may be wildly off, but these should be few and far between.

Now say you get one of those cartoony tourist maps that only show the general locations of a few landmarks. You can still make a few accurate one-to-one correspondences between your map and reality, but not many. If you try to discern much information from your map outside the relative locations of a few big landmarks, you’re going to be wildly off. In general, a pretty poor indication of reality.  
So how does this pertain to human knowledge? Well, the Perfect Map represents an omniscient entity. Everything they “know” has a perfect one-to-one correspondence with some element of reality. The entity knows that the sky is blue, and this can be mapped to the fact that the sky is blue. The entity knows all 50 states, and these facts can be mapped to the actual names of the 50 states. The entity knows the color of my underwear… well, you get the idea.

The accurate but imperfect map can be likened to someone fairly well informed, I suppose. The person knows a bunch of facts, and most of them are pretty accurate. However, if he tries to get too accurate about anything, his knowledge falls apart. He knows that it’s 3018 miles between San Fran and New York, and that’s about right. But if you ask him to express this information in nanometers, he’ll be off. He may be close, but he’ll be off. If you ask him to describe the motion of a car, he can probably tell you it’s going about 70 mph. If you try to ask him to describe the motion of every molecule within the car, he’ll probably be hopelessly off. The more specific you get in your questioning, the more off he’ll be, because you’re trying to gain information from him that’s below the scale of his knowledge base.

The tourist map can be likened to a small child, or a really dumb person, or Alec Baldwin. He’ll know his name, he’ll know what color his shirt is, and maybe how to tie his shoes, but if you try to get any details about anything, it’ll be a crap shoot. Some areas of the knowledge base may be more accurate than others through sheer chance (eg, he happened to hear the name of the capital of Vermont, so he knows this). But in general, there is a very small number of mappings you can create between his knowledge base and reality.  
And now my lunch break is over. Did I even begin to answer any of your questions? I doubt it. Was this worth a read? Eh, I wouldn’t bet money on it. At least I got to waste some bandwidth, though. 🙂 Hopefully someone will stop by who can give a more thoughtful dissertation.

Jeff

---

<div class="post-metadata">

**Author:** ![Justhink](https://avatars.discourse-cdn.com/v4/letter/j/edb3f5/32.png) [@Justhink](https://boards.straightdope.com/u/Justhink)\
**Post date:** [October 11, 2002, 3:00am UTC](https://boards.straightdope.com/t/map-and-territory-redux/129325/4 "2002-10-11T03:00:28Z")

</div>

If you map the entire data-sat of ‘rationality’; it _does_ correspond.

Asks (hypothetical, though not IMO) computer: What are next Wednesday’s lotto numbers in my reality?

Computer: 36, 12, 7, 14, 45

The types of scenarios that you are speaking of later on deal with the issue of mapping SENSES – holographic live feeds of whatever data feild you are inquiring about.

To the degree that accuracy is measured in time; a time can be inputed for a veiwing scenario so that the map is accurate when you are physically planning on encountering the scenario of inquiry.

A perfect mapping of rationality does possess perfect correspondence –

It even possesses the perfect correspondence to communicate the perfect perspective in which to veiw your own question the way you _actually_ want to veiw it; no suprizes - so as to compensate for your shortsightedness in regards to your question and your specific inquiry. What’s the point of having the lotto numbers if a meteor fragment is going to crash into your head on the way to the corner store to but it?

In the thread earlier where you pointed out this OP, you inquired three opinions of me in regards to knowledge (thx for pointing this out, I did miss it!).

This type of mapping system cannot BE the observer. The observer is still limited by the lack of processing speed required to abstract themselves, in terms of holding a perfect data field within themselves. Omniscience = catatonia in my book; however that data processing is still occurring and still tappable to others who are aware of themselves.

That however, does not mean that a perfect moral system cannot be mapped, extracted ‘excavated’ and committed to memory; so that each individual is aware of what the perfect ethical square deck is for everyone. Anything ‘enjoyed’ outside of these parameters must be experienced virtually, so as to not impact the abstractions of ‘real’ selves existentially - this is very key, even in situations of infinite life. If the entire rationality field is going to be extracted; the process of rationality that _allowed_ this extraction to occur needs to be honored as the guiding princile of the life that USES anything for themselves under the veil of rationality; as this is the WHOLLY mapped system of rationality itself. If you want to turn your light on in your room, you need to connect the external power source to the filiment. In the same way, if you want to use rationality - show it to others and observe it for yourself, you are required to actually DO something that is very narrowly defined.

This ‘site’ of rationality to be excavated is not everything that exists; but it is everything which allows existence to be comprehended. We know that people cannot know ALL THINGS at the same time without being immobile in a ‘self’ propelled sense. This is key. From this we can infer a physical evidence of anyone who is lying about their claim of rationality, or honesty in an existential sense that goes right down to the observation of difference itself. You’re either acting a certain way _or_ you should be dead (unable to execute the actual task you claimed to have executed). This is where rationality comes from… knowing that everything else except for it, kills it; makes it’s environment to survive non-existent.

It is to say, that if one uses language to achieve a means, the language itself must be used in the precise means with which it was allowed to come into being; or else the being negates themselves existentially - and is defined as ‘suicide that doesn’t commit suicide’. ‘A possessor who has no claim.’

Rationality, is, as you articulated; more-or-less the attempt to figure out variations of permanence found in ‘chaos’.  
“Why is the lifecycle of the milk taste higher in me than these other foods? I can still tase the milk in me days later, when I wake up in the morning… I must have a vulnerability here in terms of my own permanence; what’s up with that?”

Irrationality (from the perspective of rationality) is the ability to possess something while declaring that rationality is not possible, or is non-existent (located through logical inconsistencies that negate the rational actions required to behold a degree of permanence).

Hmm… I’m rambling again. I missed your cut of that quote you selected from me, which you said shocked you; it is blank. Is that the area where I left it blank; in describing how a logical application of stasis is _rationally_ applied?

-Justhink

---

<div class="post-metadata">

**Author:** ![erislover](https://avatars.discourse-cdn.com/v4/letter/e/71e660/32.png) [@erislover](https://boards.straightdope.com/u/erislover)\
**Post date:** [October 11, 2002, 1:12pm UTC](https://boards.straightdope.com/t/map-and-territory-redux/129325/5 "2002-10-11T13:12:44Z")

</div>

The blank “quote” was intriguing only because it stated something I think mapped (;)) to this thread: that we can nitpick something about every proposition. As you noted, even A=A is not _literally_ true (as one is on the left side of the = and the other is on the right).

The only way to avoid an error of any sort is to not say anything. And the only way to let one’s self speak in the face of skepticism is to try to hide behind a theory of meaning.

If the map is not the territory, then in what sense can we say we know something?

> [@](#):
>
> It is to say, that if one uses language to achieve a means, the language itself must be used in the precise means with which it was allowed to come into being…

Well, here we agree. In fact, in most cases I think our language only allows us to say just slightly more than can be said. But mustn’t it do this if language is to be able to change over time to match a dynamic reality? Each word is like a room with a wall missing. Thousands of years ago it was a small shack with an open wall. Now it is like a maze of paths leading in every direction. Becuase there is always (metaphorically) an open border, a word can be used nonsensically; that is, it can be used in a manner that doesn’t fit it but (because of the open wall) doesn’t _not_ fit it, either. In this realm a word is sensible nonsense.

Contrariwise, because there is no “perfect” border, we can never “really” tell if a word is being used absolutely correctly because the walls are never complete. Who knows, perhaps in the future the border will close and the word will have a clear scope (like in the case of Michaelson and Morely’s experiments on the æther). But even then it is not quite closed; a word like æther still finds clear poetic use.

Perhaps it is like trying to build a structure in a huge multi-dimensonal space?

**ElJeffe** here says,

> [@](#):
>
> How does this compare with the Perfect Map? You can still construct a one-to-one correspondence between elements, but you can no longer do so to the same degree.

But to this I must note: your act of measuring to check correspondence is itself the instantiation and use of another map. If you, say, wanted to test your “map of estimating perfection” you’d also find it imperfect. It is thus the extreme act of of faith to say perfection is a limit we are approaching in our map-building. There is no guarantee of convergence!

Contrast this with **Justhink** ’s

> [@](#):
>
> If you map the entire data-sat of ‘rationality’; it _does_ correspond.

Spoken like a true rationalist. 🙂

Perhaps I should say that epistemology isn’t the _study_ of knowledge, but a metaphor for it? Would that resolve the issue of “never-being-perfect”?

---

<div class="post-metadata">

**Author:** ![other-wise](https://avatars.discourse-cdn.com/v4/letter/o/4bbf92/32.png) [@other-wise](https://boards.straightdope.com/u/other-wise)\
**Post date:** [October 14, 2002, 1:18am UTC](https://boards.straightdope.com/t/map-and-territory-redux/129325/6 "2002-10-14T01:18:24Z")

</div>

Erislover -

I don’t think the following is a hijack, but it might qualify as a, um, “sidejack”. You’re fighting the good fight here, and if my post just muddies the water, feel free to pat me on the head, give me a lolly, and send me on my way.

> [@](#):
>
> “The map is not the territory.” Fine. So what is it? (_snip_) A map is a map _of_ something… but what is this “of” consist in?

I can’t help with the “of” part, but I can rant rhapsodic about the “something” part.

In order to have a representation (a map) of something, there must first be some “thing” to represent. It seems to me that there aren’t any _things_ “out there” at all; we sort of pull them out of some _existent_ process into _non_-existent “isolation”.  
Put a single cell on a microscope slide. The cell is a “thing”. But it’s a “thing” only and precisely because you’ve “separated” it from a larger process. Zoom in on the cell and you perceive a nucleus. A nucleus is a “thing”… only because you’ve separated it out from the larger process of the cell itself. Zoom in farther, and you perceive chromosomes. Chromosomes are things because… etc.

Or, if you prefer, you can zoom out: this particular cell was isolated from a fovea, which was isolated from a retina, which was isolated… etc.

This happens with _any_ and _every_"thing": leaf, bowling ball, molecule, continent - you name it. Whether the process is sub-atomic, cosmic or somewhere in between, any given “thing” has implicitly been isolated from _some_ process, whether that separation was done physically _or by reference_. And note that “thing-creation” is arbitrary and not necessarily hierarchical: you can “separate-out” using whatever scale or _context_ you please. All of which would seem to indicate that there **are no** “things-in-themselves” (or, to paraphrase Korzybski, “The map _is_ the territory, but the territory is not the territory” 😃 ).

So, to me, a more fundamental question than “How is a reference related to the thing it’s referring to?” is: “Exactly how the hell do we end up with “things” and “processes” in the first place?”

And, if you haven’t yet scrolled away in disgust, I’d also like to point out two odd characteristics in the way we use the _words_ “thing” and “process”.

First, we frequently, bizarrely, speak of a process as if it were a thing: Q: “What are a few of your favorite things?” A: “Dude, drugs and sex, of course!” (Drug = thing. Sex = process.)

Second, our linguistic use of a thing or process tends to be non-specific: You can read a monograph on a thing called dimethyltryptamine, but the drug molecule the paper is referring to is not a _specific_ molecule in a specific body, or syringe, or weighboat. By the same token, you can talk about sex for hours… creatively and in great detail… without once referring to a _specific_ act or participant(s).

So, a “thing” refers to an unspecified process, and a “process” refers to unspecified things.

If this didn’t help, erislover, I hope it at least entertained.  
other-wise

---

<div class="post-metadata">

**Author:** ![erislover](https://avatars.discourse-cdn.com/v4/letter/e/71e660/32.png) [@erislover](https://boards.straightdope.com/u/erislover)\
**Post date:** [October 14, 2002, 12:44pm UTC](https://boards.straightdope.com/t/map-and-territory-redux/129325/7 "2002-10-14T12:44:43Z")

</div>

**other-wise** , that does not bode well for the law of identity, now, does it? 🙂

It seems to me that your thrust is that our naming conventions and groupings are more or less arbitrary. But is that really so?

---

<div class="post-metadata">

**Author:** ![Duck\_Duck\_Goose](https://avatars.discourse-cdn.com/v4/letter/d/50afbb/32.png) [@Duck\_Duck\_Goose](https://boards.straightdope.com/u/Duck_Duck_Goose)\
**Post date:** [October 14, 2002, 12:54pm UTC](https://boards.straightdope.com/t/map-and-territory-redux/129325/8 "2002-10-14T12:54:14Z")

</div>

\<\< Princess graciously drops by to repeat link for thread archives \>\>

[http://www.philosphere.com/Korzybski.html](http://www.philosphere.com/Korzybski.html)

> [@](#):
>
> What Korzybski, a Polish count and mathematician, accomplished is similar to what Aristotle did around 500 BC, and what Francis Bacon did in the 17th century. They summarised and compiled the methods of thinking up to their times. Korzybski did the same for the 20th century in the 1930’s. He wrote only two books. The first, ‘The Manhood of Humanity’ (1921) proposed seeing humans as time-binders, while plants were chemical binders and animals space binders. But it was his next book in 1933 that was most influential, ‘Science and Sanity: An Introduction to non-Aristotelian Systems and General Semantics’. S&S was listed as one of the most influential books of the 20th century by a survey of most of the prominent scientists around 1950. Gaston Bachelard, the great French scientist, philosopher and poet wrote in The Philosophy of No: ‘The psychological and even physiological conditions of a non-Aristotelian logic have been resolutely faced in the great work of Count Alfred Korzybski, Science and Sanity. This volume of almost 800 pages is the prelude to an encyclopaedia whose plan aims at the reform, in a non-Aristotelian direction, of a great many sciences. It proposes this reform as a plan of sanity, as an education of energy and as the integration of active thought into life-progress.’
> 
> Much of 'Science and Sanity ’ has been incorporated into our culture without any direct reference to it. Most people remember Korzybski’s phrase ‘The map is not the territory; the map doesn’t cover all of the territory; and the map is self-reflexive (it becomes part of the territory)’.

\<\< still has NO clue what’s going on \>\>

Carry on, all you wonderful little people…  
😃

---

<div class="post-metadata">

**Author:** ![other-wise](https://avatars.discourse-cdn.com/v4/letter/o/4bbf92/32.png) [@other-wise](https://boards.straightdope.com/u/other-wise)\
**Post date:** [October 14, 2002, 2:16pm UTC](https://boards.straightdope.com/t/map-and-territory-redux/129325/9 "2002-10-14T14:16:15Z")

</div>

Elr-  
Well, my take on the law of identity is that it’s a law about _us_, about how the way we perceive reality ends up _creating_ identity, instead of a law about how “reality” actually _is_ .

> [@](#):
>
> “It seems to me that your thrust is that our naming conventions and groupings are more or less arbitrary. But is that really so?”

Yup, at least, I think so (just to clarify; my thrust is that the _act_ of naming or grouping is what I consider to be arbitrary, not our conventions for utilizing names and groups once a context for them has been established {imposed?}).

Let’s say I waltz into the room where I keep my big pile of toys. I could start “separating” what I find there into _things_, naming some “toys” and some “non-toys” (windows, floor, me, etc) and then, I could group the toys together. I could further separate the metal toys from the plastic toys. Or the square toys from the round toys and the marbles from the Ping-Pong balls (that is, if I still had marbles: it should be obvious by this post that I’ve lost them). I could separate them by color. Or size. Or aesthetic appeal. Or date of acquisition. There is no “correct” way to separate/name/group things because all of _that_ is determined by whatever context the separating/naming/grouping occurs _within_.

So, how is naming and grouping is **not** arbitrary?

---

<div class="post-metadata">

**Author:** ![erislover](https://avatars.discourse-cdn.com/v4/letter/e/71e660/32.png) [@erislover](https://boards.straightdope.com/u/erislover)\
**Post date:** [October 14, 2002, 3:17pm UTC](https://boards.straightdope.com/t/map-and-territory-redux/129325/10 "2002-10-14T15:17:50Z")

</div>

Well, let’s look at this. You rightly say you can group things by color, aesthetic appeal, size, shape, et cetera, and then say no one grouping is more correct than another. But what is the motivation for this? That no one grouping is sufficient to say all that there is to be said about a thing? Not quite the same thing as saying that this grouping is arbitrary.

I might say, for example, that the word “arbitrary” is arbitrarily formed in that the letters a, r, b, i, …, y are not necessary to the concept of “Arbitrariness”. I might as well use the “word” “flurblicious” to mean “arbitrary”. But is there something to arbitrariness itself?

You can relate objects by size, and so we have names and measures for sizing. Perhaps where we choose to draw the line for “Big”, “small”, and “inch” is arbitrary, but is classifying things on size itself arbitrary? That is the “of” you dodged in your sidetrack (which I don’t think it is, if that eases your mind :)).

A map classifies streets as a two-dimensional abstraction. This is arbitrary; we might as well (though it may be impractical) build small models more or less to scale and examine those, or take several overhead picutres and place a few labels on them, and so on. But there is not a question in a map’s _use_, and this, I think, is _not_ arbitrary.

I like your paraphrase: the map _is_ the territory, the territory is not the territory. But it stares epistemology right in the face like a showdown at high noon. Do we know anything about the territory if all we have are maps? And that of course demands that we ask, how do we know this is a map?

---

<div class="post-metadata">

**Author:** ![other-wise](https://avatars.discourse-cdn.com/v4/letter/o/4bbf92/32.png) [@other-wise](https://boards.straightdope.com/u/other-wise)\
**Post date:** [October 14, 2002, 9:58pm UTC](https://boards.straightdope.com/t/map-and-territory-redux/129325/11 "2002-10-14T21:58:20Z")

</div>

> [@](#):
>
> "… You rightly say you can group things by color, aesthetic appeal, size, shape, et cetera, and then say no one grouping is more correct than another. But what is the motivation for this? That no one grouping is sufficient to say all that there is to be said about a thing? Not quite the same thing as saying that this grouping is arbitrary.

Hmm… this is where language gets slippery; I can interpret your response in a couple of different ways, so for now I’ll have to respond to what I _think_ you mean. I’m saying that “no one grouping is sufficient to say all that there is to be said about a thing” in the _exact_ same way that a “thing” is not sufficient to say all that there is to be said about the _reality_ (whatever that is) underlying the “thing”.

I’m saying that in order to group things, you have to have “things” to group, and the creation (if you will) of “things” is arbitrary - hence, “groups” are arbitrary.

(Would it help to say we “thingify” reality, then “groupify” things?… Nah, didn’t think so.)

> [@](#):
>
> “… Perhaps where we choose to draw the line for “Big”, “small”, and “inch” is arbitrary, but is classifying things on size itself arbitrary? That is the “of” you dodged in your sidetrack (which I don’t think it is, if that eases your mind :)).”

But what is size without measurement? If all our measurements (big, small, inch) are arbitrary, and size is determined by measurement, how can size itself _not_ be arbitrary?

> [@](#):
>
> "… Do we know anything about the territory if all we have are maps? And that of course demands that we ask, how do we know this is a map?

(Furrows brow, _feels_ the neurons scrabbling for purchase) OK, lemme see… it’s getting close to Halloween, so let’s use the color orange as an example. When I see orange, electromagnetic waves (a.k.a. photons) _really are_ hitting my retinas. The orange I see in a rainbow is a direct experience of a range of frequencies in the electromagnetic spectrum. The experience of orange (map) _is_ the territory (spectrum). However, the _banding_ of the colors, the _rainbow itself_ is _not_ inherent in the electromagnetic spectrum, it’s inherent in my (arbitrary) set of sensory equipment. The bands map out a rainbow the way streets map out a city. And the kicker is, the orange is to be found _only_ in the bands of the rainbow. The EMR, the orange, is real, but the banding is not; the map _is_ the territory, but the territory is not the territory (a rainbow consists of electromagnetic waves, but electromagnetic waves are not rainbows). So, yes, we do know _something_ about the territory, and we know it’s a map because… well, because it’s not the territory.

Or a weaker, but less convoluted example: You know that “is it a duck/is it a rabbit” optical illusion? You stare at the image, and it’s a duck. Then you do that mysto cognitive flip and, bingo!, it’s a rabbit. But the _entire time_ it’s just ink on paper. The pattern of photons hitting your retinas are _exactly_ the same, duck or rabbit; you _arbitrarily_ “thingify” the pattern into “duck” or “rabbit”.

---

<div class="post-metadata">

**Author:** ![ElJeffe](https://avatars.discourse-cdn.com/v4/letter/e/958977/32.png) [@ElJeffe](https://boards.straightdope.com/u/ElJeffe)\
**Post date:** [October 14, 2002, 11:06pm UTC](https://boards.straightdope.com/t/map-and-territory-redux/129325/12 "2002-10-14T23:06:42Z")

</div>

> [@](#):
>
> ## quote:
> 
> ## “… Perhaps where we choose to draw the line for “Big”, “small”, and “inch” is arbitrary, but is classifying things on size itself arbitrary? That is the “of” you dodged in your sidetrack (which I don’t think it is, if that eases your mind ).”
> 
> But what is size without measurement? If all our measurements (big, small, inch) are arbitrary, and size is determined by measurement, how can size itself not be arbitrary?

Size, itself, isn’t arbitrary. However, many of our words used to describe size _are_. For example, “big” by itself is meaningless, absent context. A dust mite is big compared to a molecule, small compared to a dig. However, comparitive terms for size _aren’t_ arbitrary. “Bigger” has a precise definition, under the right circumstances, though in order to have an absolute meaning, you need to specify parameters.

You can say, for example, “a Buick Skylark is bigger than a Ford Escort.” This has no easily discernable, concrete meaning. Bigger in what sense? Length? Width? Trunk space? Cost? But if you say, “in terms of length, a Skylark is bigger than an Escort”, now you have a precise meaning, that is non-debateable in its accuracy. It’s either true, or not, but two rational people cannot come to different conclusions.

So essentially, “big”, “small”, etc, have no real meaning, any more than “pretty” or “fun” - it’s all subjective. However, “bigger”, “smaller”, and the like, _do_ have real meanings.

> [@](#):
>
> Well, let’s look at this. You rightly say you can group things by color, aesthetic appeal, size, shape, et cetera, and then say no one grouping is more correct than another. But what is the motivation for this? That no one grouping is sufficient to say all that there is to be said about a thing? Not quite the same thing as saying that this grouping is arbitrary.

I agree that no grouping is more correct than another - any random grouping is “correct” in that it is, in fact, a group. Even the null set is a “correct” group. However, some groups may be more useful. I can form 26 groups of all the people in the US based on the first letter of their first name, but that doesn’t tell me much of use about these people. But if I form 26 groups based on, say, income, now we have some information that most would consider to be useful. Of course, even groupings based on first names are useful, if, for instance, you’re performing a study to see what people name their kids.

Basically, it all boils down to context. Any fact, any piece of information, is worthless absent context. And different facts are of different value to different persons. A chef doesn’t care what the chemical formula for basil is, and a chemist doesn’t care what H2SO4 tastes like.

But just because facts, in and of themselves, are meaningless, doesn’t mean that facts in general are meaningless. Facts provide info about reality, and help us to understand it. And while the exact configuration of letters in the word “ambiguous” is unimportant (except to the extent that it allows us to communicate the idea behind the word to other people" the word does have meaning, and its existence has value. Similarly, if your map is drawn on parchment, vellum, or a giant sheet of pasta isn’t as important as the fact that it’s a legitimate representation of some element of terrain, inasmuch as it communictes information to the person looking at it.

Bottom line: information isn’t arbitrary, even if the medium the information is encoded in _is_.

Jeff

---

<div class="post-metadata">

**Author:** ![ElJeffe](https://avatars.discourse-cdn.com/v4/letter/e/958977/32.png) [@ElJeffe](https://boards.straightdope.com/u/ElJeffe)\
**Post date:** [October 14, 2002, 11:13pm UTC](https://boards.straightdope.com/t/map-and-territory-redux/129325/13 "2002-10-14T23:13:23Z")

</div>

> [@](#):
>
> ## quote:
> 
> ## “… Perhaps where we choose to draw the line for “Big”, “small”, and “inch” is arbitrary, but is classifying things on size itself arbitrary? That is the “of” you dodged in your sidetrack (which I don’t think it is, if that eases your mind ).”
> 
> But what is size without measurement? If all our measurements (big, small, inch) are arbitrary, and size is determined by measurement, how can size itself not be arbitrary?

Size, itself, isn’t arbitrary. However, many of our words used to describe size _are_. For example, “big” by itself is meaningless, absent context. A dust mite is big compared to a molecule, small compared to a dig. However, comparitive terms for size _aren’t_ arbitrary. “Bigger” has a precise definition, under the right circumstances, though in order to have an absolute meaning, you need to specify parameters.

You can say, for example, “a Buick Skylark is bigger than a Ford Escort.” This has no easily discernable, concrete meaning. Bigger in what sense? Length? Width? Trunk space? Cost? But if you say, “in terms of length, a Skylark is bigger than an Escort”, now you have a precise meaning, that is non-debateable in its accuracy. It’s either true, or not, but two rational people cannot come to different conclusions.

So essentially, “big”, “small”, etc, have no real meaning, any more than “pretty” or “fun” - it’s all subjective. However, “bigger”, “smaller”, and the like, _do_ have real meanings.

> [@](#):
>
> Well, let’s look at this. You rightly say you can group things by color, aesthetic appeal, size, shape, et cetera, and then say no one grouping is more correct than another. But what is the motivation for this? That no one grouping is sufficient to say all that there is to be said about a thing? Not quite the same thing as saying that this grouping is arbitrary.

I agree that no grouping is more correct than another - any random grouping is “correct” in that it is, in fact, a group. Even the null set is a “correct” group. However, some groups may be more useful. I can form 26 groups of all the people in the US based on the first letter of their first name, but that doesn’t tell me much of use about these people. But if I form 26 groups based on, say, income, now we have some information that most would consider to be useful. Of course, even groupings based on first names are useful, if, for instance, you’re performing a study to see what people name their kids.

Basically, it all boils down to context. Any fact, any piece of information, is worthless absent context. And different facts are of different value to different persons. A chef doesn’t care what the chemical formula for basil is, and a chemist doesn’t care what H2SO4 tastes like.

But just because facts, in and of themselves, are meaningless, doesn’t mean that facts in general are meaningless. Facts provide info about reality, and help us to understand it. And while the exact configuration of letters in the word “ambiguous” is unimportant (except to the extent that it allows us to communicate the idea behind the word to other people" the word does have meaning, and its existence has value. Similarly, if your map is drawn on parchment, vellum, or a giant sheet of pasta isn’t as important as the fact that it’s a legitimate representation of some element of terrain, inasmuch as it communictes information to the person looking at it.

Bottom line: information isn’t arbitrary, even if the medium the information is encoded in _is_.

Jeff

---

<div class="post-metadata">

**Author:** ![other-wise](https://avatars.discourse-cdn.com/v4/letter/o/4bbf92/32.png) [@other-wise](https://boards.straightdope.com/u/other-wise)\
**Post date:** [October 15, 2002, 2:53pm UTC](https://boards.straightdope.com/t/map-and-territory-redux/129325/14 "2002-10-15T14:53:12Z")

</div>

Hiya Jeff,

> [@](#):
>
> “…For example, “big” by itself is meaningless, absent context. (_snip_) So essentially, “big”, “small”, etc, have no real meaning, any more than “pretty” or “fun” - it’s all subjective. However, “bigger”, “smaller”, and the like, _do_ have real meanings.”

But “bigger” is meaningless absent context too, as you yourself have pointed out:

> [@](#):
>
> "“Bigger” has a precise definition, under the right circumstances, though in order to have an absolute meaning, you need to specify parameters.

I agree with you that “basically, it all boils down to context”; and I would have no objection to modifying my earlier statement to: "… my thrust is that the act of naming or grouping (or **measuring** ) is what I consider to be arbitrary, not our conventions for utilizing names and groups (and **measurements** ) once a context for them has been established {imposed?}).

> [@](#):
>
> “Any fact, any piece of information, is worthless absent context. (_snip_) Bottom line: information isn’t arbitrary, even if the medium the information is encoded in _is_”.

I don’t really disagree with that (although I don’t think that information resides solely in the medium), I’m just trying to understand exactly what the hell information _is_ and how we… um…er… “do it”. How do we go about establishing the contexts necessary to decode information if we need information to establish a context? The famous “Information is any difference that makes a difference” is an elegant description but a crappy explanation.

---

<div class="post-metadata">

**Author:** ![erislover](https://avatars.discourse-cdn.com/v4/letter/e/71e660/32.png) [@erislover](https://boards.straightdope.com/u/erislover)\
**Post date:** [October 15, 2002, 3:29pm UTC](https://boards.straightdope.com/t/map-and-territory-redux/129325/15 "2002-10-15T15:29:26Z")

</div>

The picture which “flips” between two _seemings_: first a duck, then a rabbit. Having seen both I can say, “Now I am seeing it as a duck.” Having _only_ seen the duck, can I still say, “Now I am seeing it as a duck?” What sense would that make?

We don’t see forks _as_ forks. But this is just what the separation between map and territory demands: “I am seeing this as a representation of San Fransisco.” Then move on _into_ the city limits and say, “I am seeing this as a street in San Fransisco.”

But, as with forks, we don’t see streets _as_ streets. So why the propensity to separate our concept from what they refer to? Especially in the case of a painting, or a fork, or something else that is man-made. Which isn’t to say a word’s meaning is what it refers to (in an is-of-identity sense) but rather to say that if we force ourselves to consider each representation _as_ a representation, then “representation” starts to become meaningless… there is nothing for it to represent!

---

<div class="post-metadata">

**Author:** ![ElJeffe](https://avatars.discourse-cdn.com/v4/letter/e/958977/32.png) [@ElJeffe](https://boards.straightdope.com/u/ElJeffe)\
**Post date:** [October 15, 2002, 3:49pm UTC](https://boards.straightdope.com/t/map-and-territory-redux/129325/16 "2002-10-15T15:49:26Z")

</div>

> [@](#):
>
> How do we go about establishing the contexts necessary to decode information if we need information to establish a context?

Okay, now _here’s_ an interesting question. It seems like a neverending circle that would be impossible to decode. We need information to establish a context, but we need context to make use of information. For which we need more information. For which we need context. And so on, ad infinitum. It’s kinda like trying to learn a foreign language by just reading a foreign dictionary. You look up a word, but you don’t comprehend the definition of course, so you look up all the words in _that_ definition, none of which you understand, so you look up… You get the idea.

It seems what we need is some initial context. Something absolute, something inherent. It would seem such absolutes exist, though, or else no infant would never be able to learn to speak - they would having nothing with which to associate the words.

I think that people are born with a certain degree of context encoded within them. When a baby sees a tree, they may not know what they’re looking at is something we call “tree”, or even what the heck the purpose of the tree is, but they can glean some extremely basic information just looking at it. They can tell that it exists, for one thing, which is important. They have a built-in mechanism that equates seeing a thing to that thing being real. They haven’t yet learned that some things they see aren’t real, of course, which is why TV and movies can be so scary and confusing. Also, and more interestingly, they haven’t learned that what they can’t see _does_ exist. When they can’t see mommy, as far as they know, mommy isn’t there. That’s why children can get so freaked out when mommy goes away - they don’t know if she’s really coming back, or even if she still is, so to speak.

But that raises an interesting point - baby sees mommy, mommy leaves, mommy comes back. Baby learns the concept of permanence - that things don’t cease to exist just because you can’t see them. But this isn’t really something that can be proven. Right now, I can’t see the room behind me. For all I know, there may be no room. There may be just a yawning abyss of nothing, that suddenly fills in whenI turn around. [turns around] Yup, there it is. I assume that it stays there, based on reason, experience, and what I’ve been told by others. Science works in largely the same way. We develop theories that sound good, and use those to explain phenomena, even when we can’t prove the theories. And since theories are proven wrong all the time, we have great precedence for believing that what we really believe about the machinations of the universe are fundamentally wrong. Yet we still continue to develop theories to explain stuff, and then assume they’re true. Why? Because we like to make sense of stuff.

Humans are designed to fill in the cracks. Our mind perceives various distinct events, and tries to find links between them. Even our eyes look at individual still frames of a movie - or even of reality - and interpolate to create the illusion of motion. We try to gain information, and where we can’t gain it, we make it up.

But I’ve gone off on a wild tangent there. I was talking about inherent context. Things like the assumption that what you see really exists. Or the concept of relative size. You see something bigger than you, and you know that it’s bigger than you. The concept of size has inherent meaning. Heat has inherent meaning. Pain has inherent meaning. And one biggy - math.

On a basic level, people have a concept of numbers. You see three objects, you know that there are three objects. And this is something that requires no context. “Three” is an absolute, in and of itself. Addition is absolute. This isn’t to say that everyone is born with a working knowledge of higher mathematics - no one is born knowing how to take an integral - but the existence of the integral is an absolute, and can’t be changed or debated. But the important stuff, I think, are the basics. The stuff we’re born with.

So I think that kind of addresses the question. The context/information chain isn’t endless. It goes down to some very fundamentally basic elements, from which we gradually build. But the tower of knowledge we build is fairly fragile, given that it’s based on a very small cornerstone of knowledge. If anyone of our basic, inherent, assumptions are off, then - well, we’re kinda screwed, aren’t we? 🙂

Jeff

---

<div class="post-metadata">

**Author:** ![other-wise](https://avatars.discourse-cdn.com/v4/letter/o/4bbf92/32.png) [@other-wise](https://boards.straightdope.com/u/other-wise)\
**Post date:** [October 15, 2002, 4:45pm UTC](https://boards.straightdope.com/t/map-and-territory-redux/129325/17 "2002-10-15T16:45:43Z")

</div>

Erislover,

> [@](#):
>
> “Having only seen the duck, can I still say, “Now I am seeing it as a duck?” What sense would that make?”

Damn… took me awhile to parse that; language starts getting pretty thin out here at it’s edges.  
I’m sorry, Eris, but I may be missing your point. I _think_ you’re saying that if you never “see” the rabbit, it would be nonsensical to think of it as anything other than a picture of a duck. But the fact that you’re not cognizant of _both_ representations doesn’t alter the fact that two very different representations were available, and in just _one_ pattern of “reflected” photons. Which is why I’m suggesting that we (for lack of a better term) create “things” out of an otherwise undifferentiated reality, and then act as if the “things” _are_ reality. That doesn’t mean that “things” aren’t real; reality is what they’re _made_ of. But there isn’t any “real”, most preferred, way to slice up reality - the differentiation is arbitrary (granted, we happen to have physical (and other) constraints in divvying up reality, but differentiation is still arbitrary _in principle_).

> [@](#):
>
> “… but rather to say that if we force ourselves to consider each representation _as_ a representation, then “representation” starts to become meaningless… there is nothing for it to represent!”

But aren’t representations _inherently_ meaningless in the sense that anything can be used to represent anything else, given the right context? Sure, we don’t see streets _as_ streets, but is there really any reason besides convenience _not_ to?  
Eris, I can _glimpse_ what you’re saying here, but I can’t grasp it. Could you thread this out a little more?

---

<div class="post-metadata">

**Author:** ![erislover](https://avatars.discourse-cdn.com/v4/letter/e/71e660/32.png) [@erislover](https://boards.straightdope.com/u/erislover)\
**Post date:** [October 15, 2002, 5:07pm UTC](https://boards.straightdope.com/t/map-and-territory-redux/129325/18 "2002-10-15T17:07:43Z")

</div>

> [@](#):
>
> But aren’t representations inherently meaningless in the sense that anything can be used to represent anything else, given the right context?

But doesn’t that just push off the question? I mean, what context?

“This is a sentence.” What sentence?

“This sentence makes sense.” What sense?

> [@](#):
>
> Sure, we don’t see streets as streets, but is there really any reason besides convenience not to?

That’s _almost_ the question! 🙂 Somewhere in all of our representations _must be_ a representation that is also the thing it represents (a pure intermediary?). Otherwise there is nowhere for the representation to attach to reality; that is, otherwise, it isn’t actually _representing_ (for what could the verb “to represent” mean? How would we learn its use?).

**erl** (pointing): “This is a representation.”  
**other-wise** : “What is it representing?”  
**erl** (pointing somewhere else): “Why, _this_.”

And can we raise the objection here, “Yes, but your mental grasp of that is itself only a representation of it”? No! Because then the dialog stands on its head and _I_ ask _you_ the very question _you_ just asked _me_: “What is it representing?” And if each of us has asked the question of the other, what could any answer either of us gave hope to demonstrate?

Dear god, I think I am becoming an idealist.

---

<div class="post-metadata">

**Author:** ![other-wise](https://avatars.discourse-cdn.com/v4/letter/o/4bbf92/32.png) [@other-wise](https://boards.straightdope.com/u/other-wise)\
**Post date:** [October 15, 2002, 5:13pm UTC](https://boards.straightdope.com/t/map-and-territory-redux/129325/19 "2002-10-15T17:13:24Z")

</div>

Jeff-

> [@](#):
>
> “… You see three objects, you know that there are three objects. And this is something that requires no context. “Three” is an absolute, in and of itself.”

“Three” requires the context “knowledge of objects”. If you can’t differentiate objects, you can’t know there’s three of them (but, dammit, there _is_ something inherent about math, isn’t there? I need to re-read Lakoff).

And you’re right: If there’s no initial/inherent context, it’s a circle. But how could it be a circle? It’s like picking up a board you’re standing on. But if there _is_ inherent context, how do we acquire it?

---

<div class="post-metadata">

**Author:** ![other-wise](https://avatars.discourse-cdn.com/v4/letter/o/4bbf92/32.png) [@other-wise](https://boards.straightdope.com/u/other-wise)\
**Post date:** [October 15, 2002, 6:16pm UTC](https://boards.straightdope.com/t/map-and-territory-redux/129325/20 "2002-10-15T18:16:32Z")

</div>

Erislover -

(bolding mine)

> [@](#):
>
> "… That’s _almost_ the question! Somewhere in all of our representations _must be_ a representation that is also the thing it represents (a pure intermediary?). Otherwise **there is nowhere for the representation to attach to reality** ; that is, otherwise, it isn’t actually _representing_ (for what could the verb “to represent” mean? How would we learn its use?). "

But that’s part of what I’m trying to say; representations (and “things”) are _already_ **de** -tached (distinguished/separated-out/objectified) from reality. THAT’S what representations (and “things”) _are_. Which is probably why they can be arbitrary. They’re not separated as in “floating loose in space”, they’re separated as in “ignore everything but this”. A representation isn’t so much a pointer _to_ something, as it is a pointer _away from_ everything else. When you point to a representation, or to what it represents, you’re still not telling me what it _is_, you’re just telling me what it _isn’t_.

You know, what I just wrote makes sense, but I am no longer able to tell you why.

[Next page](https://boards.straightdope.com/t/map-and-territory-redux/129325.md?page=2)
