Chapter 15 - Amendments

91    Putting of amendments

  1. In respect of every amendment the President shall put a question – That the amendment be agreed to.
  2. An amendment proposed shall be disposed of before another amendment to the original question may be moved.
  3. An amendment proposed may be withdrawn by leave of the Senate.
  4. When amendments have been made the main question shall be put as amended.
  5. When amendments have been proposed but not made the question shall be put as originally proposed.

Amendment history

Adopted: 19 August 1903 as SOs 139, 140, 143 and 144 (corresponding to paragraphs (2) to (5))

Amended: [13 May 1987, J.1849–50, paragraph (1) adopted as a sessional order in place of old SOs 141 to 145 and 149]

1989 revision: Old SOs 146, 147, 150 and 151 together with the 1987 sessional order combined into one, structured as five paragraphs and renumbered as SO 91 ; minor simplification of language


Standing order 91 deals with the mechanics of putting the question on amendments. Paragraph (1), initially adopted as a sessional order in 1987, represented a revolution in procedure, replacing, as it did, a complicated and potentially muddled scheme for dealing with amendments. Under the old scheme, amendments were dealt with in two steps. The first step was to ascertain whether there was support for the removal of the relevant words from the motion (“…the President shall put a Question, That the words proposed to be left out be left out’”).[1] If this was agreed to, the question was then put on the words proposed to be inserted or added. If the proposed substitution was not agreed to, however, the result was known in parliamentary terminology as a “mutilated motion” which may, depending on the omissions and proposed insertions, comprise as little as the word “That”. While of “ancient usage”, the practice had nonetheless been abandoned by the House of Lords in 1952 and the House of Commons in 1967. In addition, there were complicated rules which prevented further amendments being moved once parts of a question had been agreed to.[2]

The flaw in the old scheme was highlighted by proceedings on the Australia Card Bill in 1986. This bill, which sought to introduce a national identity card, was, and remains, a highly contentious measure and created a trigger under s.57 of the Constitution for the double dissolution in July 1987. On 10 December 1986, the bill was at the second reading stage in the Senate for the first time. On an opposition amendment to the motion for the second reading (in the form “Omit all words after That’, substitute: …”), the Senate agreed to omit the words in question but failed to agree to insert other words in their place, an amendment (by the Australian Democrats) to the words proposed to be inserted having also been defeated. With the motion reduced to a meaningless “That”, the government was given leave to move the second reading motion again. This was then defeated by 32 votes to 26, a procedure which at least resulted in a clear and definite decision. The bill thus suffered its first rejection for the purposes of s.57.

The matter was considered by the Standing Orders Committee which presented a report on 30 March 1987 recommending a new and more rational procedure.[3] The committee noted that the same difficulty could occur in relation to provisions of bills because of the way the standing orders required the question to be put on amendments to bills, as well as to motions.[4] Its analysis is worth quoting at some length:

A mutilated motion or provision in a Bill arises because of the procedure whereby amendments to leave out and insert words are put in two separate questions, and the rule that the words to be inserted may not be amended until the question for leaving out words has been determined.

There would appear to be no good reason for not putting all amendments in one question: “That the amendment be agreed to”, and the only other rules which would then be required would be prohibitions on the moving of an amendment the same in substance as an amendment already determined and on any amendment that would simply reverse an amendment already made.

These proposed changes would avoid the potentially embarrassing situation referred to, and also avoid the inconvenience of putting and determining amendments in two questions.

The existing rules were inherited from the British Parliament and date from the seventeenth century. The rationale of the rules is not clear, but it appears that there may be two reasons for them. First, they avoid the situation of the Senate wasting time over determining, by amendment, the words which are to be inserted in a motion before the question of leaving out words has been determined. That question, it may be thought, ought to be determined before the Senate is asked to determine the words to be inserted, otherwise a majority which does not want to amend the motion is asked to determine an amendment in which it has no interest, The argument against this is that the Senate ought to know what the words to be inserted are before it is asked to decide whether to leave words out. Theoretically, the way in which a Senator votes on the latter question could well be determined by the nature of the words to be inserted. In order to achieve that gain, it may be preferable to run the risk of wasting time over the words to be inserted, and in any case such a waste of time is less irrational and less embarrassing than being left with the word “That” or some other mutilated motion or provision in a Bill.

Secondly, the present rules may be thought to minimise the possibility of an endless series of amendments being moved to some motion, perhaps for the purpose of obstruction. The answer to this is that the proposed “same in substance” and “non-reversal” rules[5] would impose a sufficient restriction, and that the existing rules may in any case be circumvented by a little more ingenuity. If Senators aim to obstruct they always have means of doing so.

The balance of argument, therefore, would appear to lie clearly with the proposed change.[6]

The committee’s recommended solution, adopted as a sessional order on 13 May 1987, was re-adopted in the following Parliament[7] and incorporated into the 1989 revision of the standing orders on which it had had a seminal effect.

While ultimately flawed, the ancient procedures nonetheless embodied a sound principle of parliamentary procedure that is preserved in the revised standing orders. By moving amendments to motions or amendments to amendments (see SO 93) members of a deliberative assembly could be exposed simultaneously to a range of alternative propositions, but by keeping only one question before the chair at a time the assembly ensures that the final outcome represents a majority decision.[8]

For additional commentary, see Odgers’ Australian Senate Practice, 12th edition, p.185.