1, 2, 3…
There is 1 associative law.
There are 2 identity laws.
There are 3 bookkeeping laws.
–Conceptual Mathematics; pp. 21, 166 – 9
Me: So you don’t like 0! It’s not like you don’t know about 0; why you ignore 0?
You: Dang! Ok, then, let’s get even more primordial – begin with an object
•
0 is an endomap – an arrow going from • to •
• –0–> •
So is number 1 i.e. another arrow
• –1–> •
Me: Let me get this straight. The domain object of the map (number) • –0–> • is
•
and the codomain object of the number 0 is also
•
You: Yes (that’s why we call • –0–> • an endomap; Conceptual Mathematics, page 15) and so is the case with number • –1–> • i.e.
domain (1) = •
codomain (1) = •
Me: Don’t you need an identity map for each object (assuming you are going to pull a category sooner or later). Which one of the two maps
• –0–> •
• –1–> •
is the identity map of the object •?
You: First, let’s thank God (or Money for) there is only one object to worry about. Having said that, it’s tad bit tricky.
Me: I knew it—you don’t have an answer to my simple question: which one of the two numbers 0, 1 is the identity?
You: Well, when we say number 1 is identity as in
1 × 1 = 1
1 × 2 = 2
1 × 3 = 3
.
.
.
we are saying so in view of multiplication. There is something analogous to multiplication—composition—in our scenario of object • and maps
• –0–> •
• –1–> •
which we have to foveate first.
Me: Do we have to think about composition before deciding on the identity because of the two identity laws?
You: Yes; say, for illustration, the map
• –0–> •
is the identity of the object •, then the identity map has to satisfy the two identity laws which involve composition.
Given a map, say,
1: • –> •
we need the identity map
0: • –> •
of the object • to satisfy
1 + 0 = 1
and
0 + 1 = 1
with ‘+’ denoting composition of maps
• –0–> • –1–> • = • –1–> • (1 + 0 = 1)
• –1–> • –0–> • = • –1–> • (0 + 1 = 1)
Me: Since the maps
0: • –> •
1: • –> •
are numbers 0, 1, respectively, in your schizoid scheme, why not take addition (+) as composition of maps?
You: As you wish. As of now we have
Object (•)
Maps (0, 1)
Composition (+)
Identity laws (0 + 0 = 0, 0 + 0 = 0; 1 + 0 = 1, 0 + 1 = 1)
but not enough!
You: For each pair of composable maps we need a composite map. We say a pair of maps is composable if the codomain object of one map is same as the domain object of the other map in the pair.
Me: But we have only one object
•
which is both the domain and the codomain object.
You: Yes [as a result of your observation] we can add on and on
0 + 0 = 0
0 + 1 = 1
1 + 1 = 2
1 + 2 = 3
.
.
.
Me: K, I got it! We take natural numbers (0, 1, 2…) as maps, addition of numbers (+) as composition of maps and number 0 as the identity map. As much as I hate to sound are we there yet…
You: Yes, we have a category—a monoid.
Definition: A category with exactly one object is called a monoid (Conceptual Mathematics, page 166).
Me: Just between you and me… what good is knowing this monoid… I mean can I pawn it for a night out-downtown?
You: By the time I’m done interpreting monoid you’ll be stand up straight in attention every time you hear subtraction as if it were national anthem: God Bless Grassmann
Me: We’ll see
i’m completely stuck on this one–1
that’s how i feel when i try to read anything i wrote [which is more than a week old]… i guess the muse that made me write moves on ;(
is it possible to prove that in the category of sets there exists an isomorphism between two objects iff there exists a bijection between the two sets?
http://assets.cambridge.org/052180/4442/sample/0521804442ws.pdf