1.2 Arithmetic Operations in a Field

All arithmetic properties of real numbers can be deduced from the axioms stated earlier. Now, we explore the consequences of the first six axioms (I-VI) which holds for every every (even unordered) field \(F\).

Definition 1.2.1. Given two elements \(x\) and \(y\) of a field \(F\). we define their difference, \begin {align*} x-y = x +(-y). \end {align*}

In other words, to subtract an element \(y\) means to add its additive inverse, \(-y\).
If \(y \neq 0\), we also define the quotient of \(x\) by \(y\), \begin {align*} \frac {x}{y} = x \cdot (y^{-1}) \end {align*}

Also denoted by \(x/y\). In other words, to divide \(x\) by \(y\) means to multiply \(x\) by the reciprocal of y

Hence we have defined two new operations: subtraction and division, that are just special cases of addition and multiplication, we can apply the axioms to these new operations, in the following corollaries.

Corollary 1. The difference \(x-y\) and the quotient \(x/y\) (where \(y \neq 0\)) of two real numbers \(x\) and \(y\) are themselves real numbers. (Similarly for difference and quotient of field elements in general).

In Symbols: \begin {align*} (\forall x,y \in \mathbb {R}^1) \;\;\; (x-y )\in \mathbb {R}^1, \;\; (x/y) \in \mathbb {R}^1 \;\; (\text {the latter if $y\neq 0$}). \end {align*}

Corollary 2. If \(a,b,c\) are elements of field \(F\), with \(a=b\), then \begin {align*} a+c=b+c \;\; \text {and}\;\; ac=bc \end {align*}

(In other words, we may add one and the same element \(c\) to both sides of the equation \(a=b\); Similarly for multiplication.)

In Symbols: \begin {align*} (\forall a,b,c \in F)\;\;a=b \implies a+c =b+c\;\; \text {and}\;\; ac=bc \end {align*}

Proof. By the properties of equality, we have \(a+c = a+c\). Now, as \(a=b\), we may replace \(a\) by \(b\) on the right side. This yields \(a+c =b+c\) as required. Similarly for \(ac=bc\). □

The Converse to this corollary is the following.

Corollary 3. (Cancellation Law). If \(a,b,c\) are elements in a field \(F\), then \begin {align*} a+c = b+c \implies a=b. \end {align*}

If, further, \(c\neq 0\), then, \begin {align*} ac=bc \implies a=b. \end {align*}

(In other words, we may cancel a summand and a nonzero factor on both sides of the equation.)

Proof. Let \(a+c = b+c\). By corollary 2, we may add \((-c)\) on both sides of the equation to get \begin {align*} (a+c) +(-c) = (b+c) + (-c) \end {align*}

By Associativity (Axiom III), \begin {align*} a+[c +(-c)]=b+[c+(-c)]. \end {align*}

As \(c+(-c) =0\) (by Axiom V), we get \(a+0=b+0\), i.e., \(a=b\) by (Axiom IV); Similarly for multiplication. □

Theorem 1. Given two elements, \(a\) and \(b\), of a field \(F\), there always exists a unique element \(x\) such that \(a+x =b\); This element equals the difference \(b-a\).
(Thus \(a+x =b\) means that \(x=b-a\).)
  If, further, \(a\neq 0\), there also is a unique element \(y \in F\), with \(ay=b\); This element equals the quotient \(b/a\). (Thus \(ay=b\), \(a\neq 0\), means that \(y =b/a\).)

In symbols: \begin {align*} (\forall a,b \in F)\;(\exists ! \, x,y \in F ) \;\;a+x =b \;\; ay =b \;\; (\text {the latter if $a\neq 0$}) \end {align*}

Proof. It is easily checked that the equation \(a+x=b\) is satisfied by \(x=b-a\). In fact, we have: \begin {align*} a+x &= a+(b-a)\\ &= (b-a)+a &(\text {II})\\ &= [b+(-a)]+a \\ &= b+[(-a)+a] &(\text {III})\\ &= b+ 0 &(\text {V})\\ &= b &(\text {IV}) \end {align*}

Thus, the equation \(a+x=b\) has at least one solution for \(x\). To prove that this solution is unique, suppose that we have still another solution, \(x'\), say. Then, we obtain \(a+x=b\) and \(a+x'=b\), so that \(a+x=a+x'\) or \(x+a=x'+a\), Cancelling \(a\) (by Corollary 3), we see that \(x=x'\), so that the two solutions coincide.

For Multiplication, similarly, it can be checked that the equation \(ay=b\) is satisfied by \(y=b/a\), given \(a\neq 0\). \begin {align*} ay &=a \cdot (b\cdot a^{-1}) \\ &= (b\cdot a^{-1}) \cdot a &(\text {II})\\ &= b\cdot (a^{-1}\cdot a) &(\text {III})\\ &= b\cdot 1 & (\text {V})\\ &= b &(\text {IV}) \end {align*}

Hence, we can again conclude that the equation, \(ay=b\) has at least one solution for \(y\). For another solution \(y'\), \(ay'=b\) implies \(ay= ay'\), Canceling \(a\) (by Corollary 3), we can indeed see that \(y=y'\). Thus, both the existence and the uniqueness of the solutions have been proven. □

Theorem 1 shows that subtraction and addition are inverse operations to addition and multiplication. Now, we can move a summand or a factor from one side of the equation to another side.

Corollary 4. For any element \(x\) of a field \(F\), we have \(0-x=-x\). If, further, \(x\neq 0\), then \(1/x=x^{-1}\).

Proof. We have by definition, \begin {align*} 0-x &=0+(-x)\\ &= -x &(\text {IV}) \end {align*}

Similarly, \begin {align*} 1/x &= 1\cdot x^{-1}\\ &= x^{-1} &(\text {IV}) \end {align*} □

Corollary 5. For any element \(x\) of a field \(F\), we have \begin {align*} x\cdot 0 =0\cdot x =0 \end {align*}

(Hence we never have \(0\cdot x=1\); This is why \(0\) cannot have multiplicative inverse)

Proof. By distributivity (Axiom VI) and Axiom IV, we get \begin {align*} 0x+0x=(0+0)x =0x=0+0x \end {align*}

Thus \(0x+0x=0+0x\). Cancelling \(0x\) from both sides (Corollary 3), we get \(0\cdot x=0\), and by commutativity (Axiom II), we also get \(x\cdot 0=0\). □

Corollary 6. (Rule of signs). For any elements \(a,b\) of a field \(F\), we have

(i)
\(a(-b) = (-a)b = -(a\cdot b)\);
(ii)
\(-(-a)=a\);
(iii)
\((-a)(-b) = ab\).

Proof. Formula (i) means that \(a(-b)\), and similarly \((-a)b\), equals the additive inverse of \(ab\). Hence to prove (i), we have to show that \(a(-b) +ab=0\). By distributivity (Axiom VI), Axiom V, and Corollary 5 we have \begin {align*} a(-b) +ab = a[(-b)+b] = a \cdot 0 = 0 \end {align*}

Also, similarly, \begin {align*} (-a)b +ab = b[(-a)+a] = b\cdot 0 =0 \end {align*}

For (ii), we need to show that \(-(-a) =a\). Because, \(-a\) is the additive inverse of \(a\), we have \(a+(-a)=0\). Also, by definition, \(-(-a)\) is the additive inverse of \((-a)\), hence, \(-a+(-(-a))=0\). By Corollary 2, we can add \(a\) to both sides of the equation: \begin {align*} a+ (-a) +(-(-a))&=a+0\\ \implies 0+(-(-a))&=a+0 \quad \text {(V)}\\ \implies -(-a)&=a \quad \text {(IV)} \end {align*}

Formula (iii) can be derived from (i) and (ii), by first using (i) twice replacing \(a\) with \((-a)\): \begin {align*} (-a)(-b) = -((-a) \cdot b) = -(-(a\cdot b)) \end {align*}

Because of closure (Axiom I), \(a\cdot b \in F\), Hence Formula (ii), applies, giving \(-(-(a\cdot b)) = a \cdot b\) □