Skip to content

Mathematics · Ch 12 — Discrete Mathematics

Definitions

12.2.1

Definitions

Starting from a familiar example. Take the ordinary addition and multiplication of natural numbers, m+nm+n and m×nm\times n for m,n∈N={1,2,3,…}m,n\in\mathbb N=\{1,2,3,\ldots\}. Both operations share two features: (1) exactly two elements of N\mathbb N are processed at a time, and (2) the resulting element is again in N\mathbb N. Any operation on a non-empty set with these two features is called a binary operation (or binary composition) in abstract algebra.

Definition 12.1. An operation ∗* defined on a non-empty set SS is a binary operation on SS if:

  1. ∗* is defined for every ordered pair (a,b)∈S×S(a,b)\in S\times S, and
  2. it assigns a unique element a∗b∈Sa*b\in S to every such ordered pair. In other words, ∗* is a rule (a function/mapping) with input in the Cartesian product S×SS\times S and output in SS:

    ∗:S×S→S;∗(a,b)=a∗b∈S,*:S\times S\to S;\qquad *(a,b)=a*b\in S,

    where a∗ba*b is a single, unambiguous element. Because the output a∗ba*b is required to lie in SS itself and never outside it, we say ∗* is closed on SS, or SS is closed with respect to ∗* -- the closure property. Every binary operation automatically satisfies closure, by definition. Definition 12.2. A non-empty set on which one or more binary operations are defined is called an algebraic structure. An equivalent way to state Definition 12.1: ∀ a,b∈S, a∗b\forall\,a,b\in S,\ a*b is unique and a∗b∈Sa*b\in S. The symbol ∗* is generic -- depending on the set it may stand for +,×,−,÷+,\times,-,\div, matrix addition, matrix multiplication, and so on. Why not every "obvious" operation is binary -- and why number systems keep expanding. ++ and ×\times are binary on N\mathbb N, but −- is not: for (3,4)∈N×N(3,4)\in\mathbb N\times\mathbb N, 3−4=−1∉N3-4=-1\notin\mathbb N. So N\mathbb N is extended to Z\mathbb Z, on which −- is binary; (Z,+,×,−)(\mathbb Z,+,\times,-) is an algebraic structure. The same pattern repeats:
  • ÷\div is not binary on Z\mathbb Z: for (1,2)∈Z×Z(1,2)\in\mathbb Z\times\mathbb Z, 1÷2=12∉Z1\div2=\tfrac12\notin\mathbb Z -- forcing the extension Z→Q\mathbb Z\to\mathbb Q.
  • Division by 00 is never defined, so ÷\div is binary on Q∖{0}\mathbb Q\setminus\{0\}, not on all of Q\mathbb Q.
  • Needing roots of equations such as x2−2=0x^2-2=0 (irrational) and x2+1=0x^2+1=0 (imaginary) forces Q→R→C\mathbb Q\to\mathbb R\to\mathbb C. C\mathbb C is the biggest of these systems, properly containing N,Z,Q,R\mathbb N,\mathbb Z,\mathbb Q,\mathbb R as subsets.

Table 12.1 summarises which of +,−,×,÷+,-,\times,\div are binary on each number system:

OperationN\mathbb NZ\mathbb ZQ\mathbb QR\mathbb RC\mathbb CQ∖{0}\mathbb Q\setminus\{0\}R∖{0}\mathbb R\setminus\{0\}C∖{0}\mathbb C\setminus\{0\}
++BinaryBinaryBinaryBinaryBinaryNot BinaryNot BinaryNot Binary
−-Not BinaryBinaryBinaryBinaryBinaryNot BinaryNot BinaryNot Binary
×\timesBinaryBinaryBinaryBinaryBinaryBinaryBinaryBinary
÷\divNot BinaryNot BinaryNot BinaryNot BinaryNot BinaryBinaryBinaryBinary