Mathematical Writing - Vivaldi Franco 2014
Conjunctions - Loops of Implications
Forms of Argument
A conjunction is a statement of the type
and
.
The statements
and
are called the conjuncts. As with implications, conjunctions are not necessary presented explicitly.
A direct proof of a conjunction consists of the separate proofs of the two conjuncts, which must be differentiated clearly. It’s common practice to put the conjuncts in an ordered list, say (i) and (ii). (If there are more than two conjuncts, the list may be extended using lower-case Roman numerals: (iii), (iv), etc.) Then the proof itself should use the matching labels (i) and (ii), and in each case we should begin by stating what we intend to prove.
PROOF.
(i)
(ii)
Equality of sets is the canonical hidden conjunction:
![]()
What are the conjuncts? Two sets are equal if they have the same elements, namely every element of
is an element of
, and vice-versa.
![]()
The structure of the proof is now determined by the definition (4.19) of a subset.
PROOF.
(i)
(ii)
Another basic conjunction is the equivalence of two statements:
if and only if
.
The equivalence operator
is the conjunct of an implication and its converse—see (4.12).
PROOF.
(i)
(ii)
We could replace either part by a proof of the contrapositive implication.
The following theorem is of this type: it provides an alternative characterisation of primality.
Theorem. 3 A natural number
is prime if and only if
divides
.
This theorem says
, where
(
is prime) ![]()
(
is prime).
The outline of the proof is now clear.
PROOF.
(i)
(ii)
The adverb ’precisely’ may be used to turn a one-sided implication into an equivalence, hence a conjunction. Thus the proof of the statement
The set
is a field precisely if
is prime.
will have to be carried out in two stages.
PROOF. Let
be given.
(i)
(ii)
7.4.1 Loops of Implications
The equivalence of several statements is sometimes expressed by a chain of implications in which the last statement in the chain coincides with the first one, thus forming a loop. This type of argument combines conjunctions with implications.
The equivalence of two statements
and
, may be viewed as a loop:
![]()
More generally, we consider loops of
implications:
![]()
Proving all implications in the loops amounts to proving that all statements are equivalent, namely that
, for all
. This follows from the transitivity of the implication operator.
For example, let
be a group and
a subgroup of
. A left coset of
in
is a set
where
is an element of
. [This is a variant of the algebraic product of sets (2.21)]. A result in group theory states that if
are elements of
, then the following statements are equivalent:
· (
) ![]()
· (
) ![]()
· (
) ![]()
· (
)
.
These are usually proved in a circle, for example (
)
(
)
(
)
(
)
(
), but other arrangements are possible, e.g., (
)
(
)
(
)
(
)
(
).
In Chap. 8 we will use a loop of implications to show the equivalence of four formulations of the principle of induction.