• A. Introduction

  Very few statutes are painted onto a blank canvas. Most legislative drafters are faced with an existing picture of laws onto which their new law must carefully be placed, sometimes painting over existing laws, sometimes blending in with them. In this sense, most legislation is amending legislation as it amends the existing law already in force. This article examines how amending legislation can be drafted in such a way as to promote clarity. By clarity it is meant “the quality of being clear and easily perceived or understood”.¹xH. Xanthaki, ‘Drafting Manuals and Quality in Legislation: Positive Contribution towards Certainty in the Law or Impediment to the Necessity for Dynamism of Rules?’, Legisprudence, Vol. 4, No. 111, 2010, p. 115.
  The article examines various techniques for promoting clarity in amending legislation. It examines in particular the rarely used technique known in the United Kingdom as the Keeling schedule. A Keeling schedule can be used whenever a new statute amends an old statute. The Keeling schedule is a schedule in the new statute setting out the old statute (or part of it) as it will read after amendment by the new statute. It is so named after Edward Keeling MP who first proposed the idea. The practice was endorsed by the UK Prime Minister in 1938.²xH.C. Deb, 26 July 1938 col 338 col 2919. The argument that this article makes is that Keeling schedules are a useful tool in improving clarity in amending legislation.
  There has been little academic work on Keeling schedules. Crabbe devotes nine lines to them in a 300-page book on drafting.³xV.C. Crabbe, Legislative Drafting, Cavendish, London, 1993. Thornton does not even mention them in his work on drafting.⁴xG.C. Thornton, Legislative Drafting, Butterworths, London, 1996. Aside from a few papers in Statute Law Review, there is a similar dearth when it comes to academic articles.

• B. Clarity in Amending Legislation

  When a new law amends an old law, it can act in two dimensions. Firstly, it can set out the changes being made to the law. Secondly, it can set out what the law will be after these changes are made. The difficulty is that the further one travels in one dimension, the less one travels in the other. It is analogous to Heisenberg’s Uncertainty Principle – the more precise your knowledge is of an object’s position, the less precise your knowledge is of its speed. Unfortunately for legislative drafters, both dimensions are important and different user groups will have different interests in each of these dimensions. Hand says that “good parliamentary drafting should have two groups of users in mind, the members of parliament who debate and pass legislation, and the ultimate users, those who are affected by the legislation”.⁵xC. Hand, ‘Drafting with the User in Mind – A Look at Legislation’, SLR, Vol. 4, No. 1, 1983, p. 166.
  Take an amendment to tax law as an example. An experienced accountant already knows the existing tax regime and will only want to know what changes are being made to it – what are the new allowances, what old exemptions are being scrapped? Legislators will likewise want to know what changes are being made by a Bill before them. They aren’t voting on the entire tax system, they are only voting on the changes to the system that are proposed. However, sometimes the changes they are voting on won’t make sense unless set in the broader context of tax law – so they might also need to know what the total law will look like after the change. A business person won’t be particularly interested in the changes made by the most recent Bill, but will be more interested in a complete and comprehensive statement of tax liabilities.
  There are two main techniques by which a new law amends an old law, the textual amendment and the non-textual amendment. A textual amendment makes a direct change to the text of the old law.⁶xCrabbe, 1993, p. 137. Thornton calls this technique ‘direct and textual’.⁷xThornton, 1996, p. 405. The Civil Partnership Act 2004 is a good example. The purpose of that Act was to give civil partners exactly the same civil rights as spouses. In many places, the Act amends older laws by inserting ‘or civil partner’ in those older laws after ‘spouse’, or by inserting ‘or surviving civil partner’ after ‘widow or widower’. As can be expected, there are many pages of such amendments consequent upon the establishment of the civil partnership institution. A huge number of older statutes had to be amended.
  A non-textual amendment makes an indirect change to an older law by requiring it to be read in a different way.⁸xCrabbe, 1993, p. 137. Thornton calls this technique ‘indirect, referential and cumulative’ although this can be confusing as textual amendments can also have a cumulative effect as they are stacked on top of each other. For example, the Department for Employment and Learning Act (Northern Ireland) 2001 changed the name of the Department of Higher and Further Education, Training and Employment to the slightly less unwieldy Department for Employment and Learning. The 2001 Act provides that in any law made before the Act, a reference to the old name shall be read as a reference to the new name. This kind of non-textual amendment is common in Northern Ireland when Ministries become Departments or change their name with the rise and fall of political tides. It saves having to make a large number of mainly cosmetic consequential amendments each time the name changes. The key difference is that the non-textual amendment is a blanket amendment which does not physically change the text of the old law. Unlike the Civil Partnership Act 2004, the 2001 Act does not trawl through the statute book identifying each instance where the Department was named and substituting the new name in its place.
  According to Samuels, “textual amendment, substitution of a new section for the old section, is desirable in principle.”⁹xA. Samuels, ‘The Courts and Legal Services Act 1990: An Exercise in Drafting’, SLR, Vol. 12, No. 76, 1991, p. 77. Berry is also of the view that textual amendment is superior to non-textual amendment¹⁰xD. Berry, ‘Keeping the Statute Book up to Date: A Personal View’, CLB, Vol. 36, No. 79, 2010, p. 98. as is Thornton.¹¹xThornton, 1996, at 407. Textual amendment is by far the preferred modern method of drafting, at least in Commonwealth countries. Non-textual amendments are easier to prepare, as there is no need to actually identify older statutes with the phrase which needs to be amended. However, the reader of the old statute will have no way of knowing that it has been superseded by the new statute. However, textual amendments can sometimes make the amending statute extremely long-winded. For example, the thirty schedules of the Civil Partnership Act 2004 take up nearly 300 pages, mainly to make many textual amendments to existing statutes.
  Ilbert is of the view that non-textual amendments perform well in the first dimension, ‘for parliamentary purposes [non-textual amendment] is the most convenient, because under it every member of Parliament who knows anything of the subject learns at once the nature of the amendment proposed’,¹²xC. Ilbert, Legislative Methods and Forms, Clarendon Press, Oxford, 1901, p. 259. although Thornton strongly disagrees.¹³xThornton, 1996, p. 407.
  In my opinion, properly executed textual amendments are reasonably good at acting in the first dimension as all they do is state the changes made to the law. For example, the effect of adding ‘or civil partner’ after ‘spouse’ is quite obvious. However, extensive use of them causes a problem with the second dimension – as more and more amendments are layered on top of an old law, it becomes increasingly difficult to work out what the old law actually says. This problem is particularly pronounced where an older statute is amended on a nearly annual basis, as is the case with much criminal justice legislation in the UK.
  Complex and continual textual amendments conflict directly with the criterion of clarity in legislation. It is not a new problem. A Select Committee Report from 1875 stated

  A reference to parts of several Acts of Parliament, some of which are repealed, some amended, and others kept alive, subject to the provisions contained in the amending Bill … makes the Statute so ambiguous, so obscure, and so difficult of comprehension that the judges themselves can hardly assign a meaning to it.¹⁴xReport from the Select Committee on Acts of Parliament (1875), p. iv.

  In fact, as long ago as 1616, Sir Francis Bacon argued for reforms to reduce “statutes heaped one upon another to one clear and uniform law”.¹⁵xQuoted in Berry, 2010, p. 81. More recently, Lord Brightman complained of ‘drafting quagmires’ where “an enactment of an earlier year is amended again and again by new Acts which add words here and delete words there, repealing one subsection and substituting a new section until the reader is totally bewildered.”¹⁶xLord Brightman, ‘Drafting Quagmires’, SLR, Vol. 23, No. 1, 2002, p. 1. See also Sutcliffe v. Commissioners of Inland Revenue, (1978) 14 TC, p. 171 and legislation in ‘unintelligible form’.
  Most of the above writers cite their own personal legislative horror stories of layers of textual amendments. As I have recently drafted amendments to the Police and Criminal Evidence (Northern Ireland) Order 1984, on behalf of the Northern Ireland Law Commission, I take Article 48 of that Order as my own horror story. Set out on the following page is a table explaining the nature of the legislative amendments to Article 48. The first column sets out each paragraph (and the title) of the Article. The second column sets out the legislative provenance of that provision, for example if it is as originally enacted or whether it has been amended by subsequent legislation. This column does not take into account any amendments which have themselves been repealed or superseded by other amendments. The third column sets out the changes to that provision which are proposed by the Northern Ireland Law Commission, in its review of bail law.¹⁷xNorthern Ireland Law Commission, Report on Bail in Criminal Proceedings (NILC 14, 2012).
  Without assistance, the nature of these amendments make it practically impossible for the user to know what Article 48 actually does. If there was no tool to help, the reader would have to look at the 1984 Order whilst simultaneously also looking at the 1995 Order, the 2003 Order, the 2005 Order, the 2007 Order, the 2008 Order as well as the amendments proposed by the Law Commission.

  Table 1.Article 48 of PACE as amended

  Paragraph	Provenance	Proposed Amendments
  Title	As originally enacted	New title
  (1)	Partially amended by the Criminal Justice (NI) Order 2003	New paragraph
  (1A)	Inserted by the Criminal Justice (NI) Order 2005	Some words substituted
  (2)	Partially amended by the Police and Criminal Evidence (Amendment) (NI) Order 2007	Some words deleted
  (2A)	Inserted by the Criminal Justice (NI) Order 2003	Some words deleted
  (3)	The original paragraph was deleted and replaced by the Criminal Justice (NI) Order 2003	Deleted
  (3A)	Inserted by the Criminal Justice (NI) Order 2003	Some words inserted
  (3AA)		Inserted
  (3B)	Inserted by the Criminal Justice (NI) Order 2003	New paragraph
  (3C)	Inserted by the Criminal Justice (NI) Order 2003
  (3D)	Inserted by the Criminal Justice (NI) Order 2003 and then partially amended by the Criminal Justice (NI) Order 2008	Some words deleted Some words substituted
  (3E)	Inserted by the Criminal Justice (NI) Order 2003
  (3F)	Inserted by the Criminal Justice (NI) Order 2003	Some words substituted Some words inserted
  (3G)	Inserted by the Criminal Justice (NI) Order 2003
  (3H)	Inserted by the Criminal Justice (NI) Order 2003	Some words inserted
  (3I)		Inserted
  (3J)		Inserted
  (4)	The original paragraph was deleted and replaced by the Criminal Justice (NI) Order 2008	Deleted
  (5)	The original paragraph was deleted and replaced by the Criminal Justice (NI) Order 2008	Deleted
  (6)	Some of the original words repealed by Criminal Justice (NI) Order 2003	Deleted
  (7)	As originally enacted	Deleted
  (8)	As originally enacted	Deleted
  (9)	As originally enacted	Deleted
  (10)	As originally enacted	Deleted
  (11)	Some words inserted by the Police (Amendment) (NI) Order 1995	Deleted
  (12)	As originally enacted	Deleted
  (13)	As originally enacted	Deleted
  (14)		Inserted

• C. Tools to Enhance Clarity in Amending Legislation

  The starting point for amending legislation is the bare textual amendment. There are then a range of tools that can assist with clarity. A non-exhaustive list of those tools is set out below. This is followed by a practical example of how legislation drafted with these tools might look. The example is an amendment to the hypothetical Theft Act 2000 which increases the maximum penalty for theft from five years to seven years.

  • Consolidation with amendments.

  • Expansive textual amendments.

  • Repeal and restatement.

  • Textual amendments with explanatory material.

  • Official reprints of statutes as amended.

  • Private reprints of statutes as amended.

  • Pamphlets, guidance and information notices.

  I. Old law – Theft Act 2000

  1. (1) A person is guilty of theft if the person dishonestly appropriates property belonging to another with the intention of permanently depriving the other of it.
  (2) A person convicted of theft is liable to imprisonment for a term not exceeding 5 years.

  This is the hypothetical old law. It contains the elements of the offence of theft and the penalty for that offence.

  II. Bare Textual Amendment

  1. In section 1(2) of the Theft Act 2000 for ’5’ substitute ’7’.

  The starting point is the bare textual amendment. The reader only knows the significance of this by reading the amendment alongside the old statute.

  III. Consolidation with Amendments

  1. This Act is the Crime Act 2012.

  2. The Theft Act 2000 is repealed.

  3. (1) A person is guilty of theft if the person dishonestly appropriates property belonging to another with the intention of permanently depriving the other of it.
     (2) A person convicted of theft is liable to imprisonment for a term not exceeding 7 years.

  A pure consolidation Act repeals all previous Acts on a particular subject and restates them in a single new Act. Aside from any necessary ‘carpentry’¹⁸xF. Bennion, Bennion on Statutory Interpretation, 5th ed., Lexisnexis, London 2008, p. 605. to fit disparate Acts into one Act, it makes no substantive change to the law. Consolidation with amendments does allow for substantive change from the old Acts. It is seen by Lord Simon as the best way to deal with the problem of multiple overlapping textual amendments.¹⁹xLord Simon, ‘Statute Consolidation: Interim Techniques’, Public Law 1985, p. 352. Lord Hailsham agreed that ‘statute law revision and consolidation are both valuable ways of improving the statute book’.²⁰xLord Hailsham, ‘Addressing the Statute Law’, 4 SLR 1985, p. 7. Samuels pleads for it.²¹xA. Samuels, ‘Consolidation: A Plea’, 26 SLR 2005, p. 56. It is difficult to find anyone who disagrees with the principle of consolidation Acts. In fact, one statutory duty of the Northern Ireland Law Commission is ‘the reduction of the number of separate legislative provisions’²²xJustice (Northern Ireland) Act 2002, s. 51(1). and a similar statutory duty of the Law Commissions of England and Wales and Scotland is to prepare ‘comprehensive programmes of consolidation’.²³xLaw Commissions Act 1965, s. 3(1)(d).
  The main difficulty is a practical one. Consolidation Acts are very resource intensive, both in terms of drafting work and parliamentary time (although special procedures do exist to speed consolidation Acts through Parliament²⁴xConsolidation of Enactments (Procedure) Act 1949, Standing Orders of the House of Commons, Standing Order 58.). A secondary difficulty is the rapidity of legislative change, meaning that a consolidation Act itself can become out of date within a short time, as new Acts amend it.
  Consolidation is good in the second dimension as the reader knows what the new law is, but doesn’t perform as well in the first dimension as the reader will be hard pressed to know what changes are being made to the law.

  IV. Expansive Textual Amendment

  1. For section 1(2) of the Theft Act 2000 substitute –

  ‘(2) A person convicted of theft is liable to imprisonment for a term not exceeding 7 years.’

  Expansive textual amendment is a tool designed to enhance clarity. Rather than insert or substitute one or two words in a section, expansive textual amendments substitute or restate the entire section. This tool expressly forms part of drafting principles in the European Union – “Preference shall be given to replacing whole provisions (articles or subdivisions of articles) rather than inserting or deleting individual sentences, phrases or words”.²⁵xGuideline 18, Interinstitutional Agreement of 22 December 1998 on common guidelines for the quality of drafting of Community legislation (1999/C 73/01). Although according to Robinson, these guidelines are not binding.²⁶xW. Robinson, ‘Manuals for Drafting European Union Legislation’, 4 Legisprudence 2010, p. 129.

  V. Repeal and Restatement

  1. Section 1 of the Theft Act 2000 is repealed.

  2. (1) A person is guilty of theft if the person dishonestly appropriates property belonging to another with the intention of permanently depriving the other of it.

  (2) A person convicted of theft is liable to imprisonment for a term not exceeding 7 years.

  A related tool is repeal and restatement. With this, instead of amending or inserting a whole section into the old Act, the section in the old Act is repealed and restated in the amending Act. The House of Lords advocated this technique, “where possible, statutes or complete parts of statutes, should not be amended but re-enacted in an amended form so that those concerned can read the rules in a single document”.²⁷xMerkur Island Shipping Corporation v. Laughton, [1983] 2 AC 570, p. 570, at p. 595. Lovgren argues that where textual amendments are made to many Acts, it is better instead to make a new Act.²⁸xA.C. Lovgren, ‘The Location of Textual Amendments in Bills’, 9 SLR 1988, p. 27.
  Although both these tools can make the new section more intelligible under the second dimension, it can be criticized for failing the first dimension. If the entire provision is substituted or restated, it is difficult, without turning to extraneous material, to work out the change between the old law and the new law.

  VI. Textual Amendment with Explanatory Material

  1. In section 1(2) of the Theft Act 2000 (penalty for theft) for ‘5 years’ substitute ‘7 years’.

  Textual amendments can also be drafted in a way which gives a better indication of the effect of the amendment. This can be done by careful use of explanatory material within the legislation, generally by including references in parenthesis to the Act which is being amended. It is possible to tinker with the exact amending words to enhance the clarity of the amendment. The drafter might amend more words than are strictly necessary, but the nature of the amendment will be more intelligible to the reader. In some jurisdictions this type of explanatory material is prohibited in statutes. This tool is therefore not always available.

  VII. Reprints

  Theft Act 2000 [a]
  1. (1) A person is guilty of theft if the person dishonestly appropriates property belonging to another with the intention of permanently depriving the other of it.
  (2) A person convicted of theft is liable to imprisonment for a term not exceeding 7[b] years.
  [a] as amended by the Theft (Amendment) Act 2012
  [b] amended by s. 25(1) Theft (Amendment) Act 2012

  Samuels and Lord Simon both recommend that statutes be officially reprinted in their amended form. Therefore, if a new statute amends an old statute, the old statute should be immediately available in its amended form. If this is done by the official government printer, then it will have an imprimatur of authority, even if it doesn’t have the authority of an Act of Parliament. Beatson cites an example from Scotland where the government published an unofficial version of an Act as amended ‘because without it the new law was utterly incomprehensible’.²⁹xJ. Beatson, ‘Common law, Statute Law and Constitutional Law’, SLR, Vol. 27, No. 1, 2006, p. 4. These reprints can be ad hoc or they can be done on an annual basis. Berry surveys several jurisdictions on their approach to reprints and gives the example of the Crimes Act 1958 (Victoria) which has been reprinted 21 times since originally enacted.³⁰xBerry, 2010. A slight variation of the reprint is the public database of the statutes as amended, provided in the UK by Her Majesty’s Stationery Office. Hughes and Davies strongly recommended that all Welsh legislation be available online in its amended form as soon as possible.³¹xD. Hughes & H. Davies, ‘Accessible Bilingual Legislation for Wales’, 33 SLR 2012, p. 103.
  Reprints are an excellent tool in theory. When an amending Act is made, the user can see the nature of the changes to the law in the amending Act. If the user wants to know the law in its totality, they can refer to the old Act as amended by the new Act. The main defect is practical – the old law must be republished as soon as, and every time it is amended, and the database must be continually and immediately updated. This is a costly undertaking. The other defect can be accuracy: reprints and databases don’t have the guarantee of accuracy provided by an Act.
  Teasdale considers consolidation, repeal and restatement and reprints as all valuable tools for contributing to the quality of the statute book.³²xJ. Teasdale, ‘Statute Law Revision: Repeal, Consolidation or Something More?’ 11 Eur J L Reform 2009, p. 157. The Renton Committee likewise saw the benefit of these tools in different circumstances.³³xRenton Committee, Preparation of Legislation (1975) Cmnd 6053.
  Several commercial bodies reprint the statutes as updated, or provide subscription based access to databases of up-to-date legislation as amended. In the UK, there is Halsbury’s Statutes. Online, there is Westlaw and Lexisnexis. In practice, these are reasonably good. However, they can suffer the same problems as public reprints, viz. they must be kept up to date and there is no guarantee of accuracy. They suffer from an additional defect – they are private databases and not free to use. One pillar of the rule of law is that law must be freely accessible, not limited to those who can pay for access.

  VIII. Pamphlets

  Finally, there are pamphlets, guidance and information notes on amending legislation. These can be either official or unofficial. They can be for internal use by civil servants administering the law, or they can be publicly available. Parliament cited these as a reason why a consolidation Act was not necessary for the Social Security Fraud Bill 2000/01.³⁴x(28 March 2001) 624 HL debs. col 264. The first problem with this tool is the familiar one of lack of authority. As Brightman said, “commercial publications, leaflets, codes of guidance and computer print-outs do not possess the authority of an Act of Parliament”.³⁵xBrightman, 2002, p. 6. The second concomitant problem is lack of guarantee of accuracy. The third problem is more fundamental – ignoring legislation in favour of a pamphlet is inimical to legal authority. As has been stated “we view with considerable misgiving the widespread practice of government by circular, with no legal force. Legislation is so complex that administrators … use circulars instead of text”.³⁶xSociety of Public Teachers of Law, ‘Preparation of Legislation’, JSPTL, Vol. 13, No. 96, 1974, p. 99. We live subject to the rule of law, not the rule of pamphlets.

• D. Keeling Schedule Solution

  1. In section 1(2) of the Theft Act 2000, for ‘5’ substitute ‘7’.

  2. For convenience, Schedule 1 sets out section 1 of the Theft Act 2000 as amended by this Act.

  Schedule 1 – Section 1 of the Theft Act 2000 as amended
  1. (1) A person is guilty of theft if the person dishonestly appropriates property belonging to another with the intention of permanently depriving the other of it.
  (2) A person convicted of theft is liable to imprisonment for a term not exceeding 7 years.

  The Keeling schedule is a useful tool to solve the problem of lack of clarity in amending legislation. An Act with a Keeling schedule addresses the two dimensions of the problem. In the body of the amending Act, it sets out, by textual amendment, precisely the change being made to the old Act. The legislators are therefore fully aware of this. The first dimension is explained. Then, in the schedule, the old Act will be restated as it now has effect. Therefore, end users are clear about what the law now is. The second dimension is explained. There is an additional bonus. When considering the first dimension, the changes being made can be easily seen in the context of the old Act. The legislators can flip backwards and forwards and better understand the nature of the changes being made. The goal of clarity is attained, both for those who want to know the changes, and those who want to know the new law. The simple tool of a Keeling schedule improves the quality of legislation.
  All writers who have considered the Keeling schedule are in favour of it. The Renton Committee recommended it.³⁷xPara. 13.21. The Statute Law Society recommended it twice.³⁸xStatute Law Society, Statute Law. The Key to Clarity (1972) para. 31. Statute Law Society, Statute Law: Radical Simplification (1974) paras101-120. Every time Samuels considered it, he has recommended it.³⁹xSamuels (1974, 1991, 1997, 2005). Bates recommended it,⁴⁰xT. Bates, ‘Monitoring of Current Legislation by the Statute Law Society’, 1 SLR 1981, p. 39. as did Hand, Berry and Lord Simon. Carr considered it would improve the statute book.⁴¹xC. Carr, ‘British Isles: Legislation’, 31 J Comp Legis & Int’l L 1949, p. 1. Zander called it a ‘simple and attractive idea’.⁴²xM. Zander, The Law-Making Process, Cambridge University Press, Cambridge, 6th ed., 2004, p. 29. Lord Brightman classified it as ‘simple ... to draft, and helpful … to readers’.⁴³xBrightman, 2002, at 3. Beatson called them ‘useful’.⁴⁴xBeatson, 2006, p. 4. Lord Hailsham said “I will always do what I can to encourage … a Keeling schedule”.⁴⁵xHailsham, 1985, p. 8.
  Three further aspects of the Keeling schedule copper-fasten their appeal. Firstly, they are part of an Act of Parliament and therefore have exactly the same authority as an Act of Parliament. Secondly, they are properly drafted by legislative counsel and scrutinized by legislators, so their accuracy is the same as any other piece of legislation. Thirdly, they are proper laws and are publicly promulgated, so they comply with legal authority and the rule of law.
  Why then are Keeling schedules used so rarely? Since they were introduced in 1938, on average there has been approximately one per year. There are seven arguments against their use, which in my view can be largely disposed of.
  Firstly, they are time consuming to prepare as essentially the drafter has to draft everything twice, once in the body of the amending Act and once in the schedule. I disagree. The main work is done in preparing the amendment, the schedule only requires careful copying of the amendment into the old law. In fact this process will most likely already be done informally by the drafter in order to check that the amendment works in the old statute. As a practical point, if the policy continually changes during the drafting process, it can be difficult to constantly keep updating the Keeling schedule. For this reason, where possible, the Keeling schedule should be left to the end, after the policy has been definitively settled.
  Secondly, they are only of value if users know of their existence, and as they are buried in a schedule of an amending Act, the user may not know of them. This is true to an extent, but the problem here is more the volume of legislation overall, rather than the location of a particular Keeling schedule. Whenever the user reads the most recent amending Act, they will be directed to the Keeling schedule in that Act. Furthermore, it is certainly easier to find one Keeling schedule than to scan the statute book for every single statute that amends an old Act.
  Thirdly, the Keeling schedule will be superseded as soon as a new amending Act is made. This is also true, but is perhaps an argument in favour of slowing the pace of continual legislative amendment, rather than criticising Keeling schedules for failure to keep up with that pace. Furthermore, the new amending Act could likewise contain a new Keeling schedule.
  Fourthly, bare textual amendments conceal the nature of legislative change and make it harder for the opposition to wreck a government legislative plan. Of course, no government would ever explicitly state this, although it is the argument that Brightman puts forward on their behalf. This is an argument for complexity and against clarity and as such can be disposed of as contrary to achieving quality in legislation.
  Fifthly, a continuously updated online statute law database is better than piecemeal Keeling schedules which are only useful until the date of the next amendment. This argument holds weight in an ideal world where such databases exist. However, in their absence, the Keeling schedule does what those databases should be doing. In addition, it does this with the authority of an Act of Parliament. Furthermore the Keeling schedule assists legislators with the passage of the Bill through Parliament, something which databases cannot do with their restriction to Acts post enactment, not to Bills passing through Parliament.
  Sixthly, where a Keeling schedule covers matters which were controversial when first enacted, the Keeling schedule may give legislators a chance to re-open contentious old debates. The argument is that a bare textual amendment only gives the chance for debate on the words changed, whereas a Keeling schedule allows for debate on the old Act as a whole. This argument was fatally wounded by the Renton Report, which stated that, since the first use of Keeling schedules, parliamentary procedure was such that amendments could not be made to that schedule “except those which were strictly consequential upon amendments to the substantial provisions of the Bill.”⁴⁶xRenton, para. 13.21.
  Seventhly, they add excess length to the statute book. Bennion states that it ‘uselessly clutters up the statute book’.⁴⁷xF. Bennion, Bennion on Statute Law, 3rd ed., Longman, London, 1990. Bates argues, “when a Bill already has 265 pages and weighs nearly 1 lb, one can see that the addition of further schedules is not a light matter in any sense of the adjective”.⁴⁸xBates, 1981, p. 39. In a similar vein of argument, Keeling schedules lead to duplication of the statute book as the law appears in three places: the old Act, the amending provisions of the new Act, and the Keeling schedule in the new Act. This is a strong argument and there will be cases where the additional length created by a Keeling schedule cannot be justified. The drafter will need to weigh up the competing arguments here of clarity versus length and complexity. Under Xanthaki’s model of quality in legislation, the advantage of clarity trumps the disadvantage of duplication of legislation.⁴⁹xH. Xanthaki, ‘On Transferability of Legislative Solutions: The Functionality Test’, in C. Stefanou, H. Xanthaki (Eds.), Drafting Legislation: A Modern Approach, Ashgate, England, 2008. The example cited above of the Civil Partnership Act 2004 is one place where a Keeling schedule would not work. A Keeling schedule for each statute that the 2004 Act amended would extend the 2004 Act to thousands of pages. The Keeling schedule is more suited where substantial amendments are made to one statute rather than smaller number of amendments to many statutes. See for example the London Local Authorities Act 2012 which sets out a Keeling schedule to show how the City of Westminster Act 1999 looks after amendment.

• E. Conclusion

  It has been demonstrated that Keeling schedules are effective in improving the clarity of amending legislation. They do so by promoting clarity in two dimensions. Firstly they help explain the nature of changes being made to the law. Secondly, they help explain what the new law will be after those changes are made. They are as effective as the tools of consolidation and repeal and re-enactment, without quite being as resource intensive as those tools. They are more accurate and have greater authority than official reprints of legislation. As well as greater accuracy and authority, they have more legitimacy than commercial reprints of legislation and don’t undermine the rule of law in the way that government by pamphlet does. Some arguments against them are more about the convenience of the government than the clarity of the legislation. The technological solution may be the most optimum, but it is one which isn’t yet fully in operation.Keeling schedules are not a magic bullet to solve all problems of clarity in amending legislation. For example, they won’t be of much assistance where amendments are being made to multiple statutes. However, they are a useful weapon in the drafter’s armoury. It is submitted that this neglected technique should be re-invigorated and used to improve the quality of legislation.

Noten

• 1 H. Xanthaki, ‘Drafting Manuals and Quality in Legislation: Positive Contribution towards Certainty in the Law or Impediment to the Necessity for Dynamism of Rules?’, Legisprudence, Vol. 4, No. 111, 2010, p. 115.

• 2 H.C. Deb, 26 July 1938 col 338 col 2919.

• 3 V.C. Crabbe, Legislative Drafting, Cavendish, London, 1993.

• 4 G.C. Thornton, Legislative Drafting, Butterworths, London, 1996.

• 5 C. Hand, ‘Drafting with the User in Mind – A Look at Legislation’, SLR, Vol. 4, No. 1, 1983, p. 166.

• 6 Crabbe, 1993, p. 137.

• 7 Thornton, 1996, p. 405.

• 8 Crabbe, 1993, p. 137.

• 9 A. Samuels, ‘The Courts and Legal Services Act 1990: An Exercise in Drafting’, SLR, Vol. 12, No. 76, 1991, p. 77.

• 10 D. Berry, ‘Keeping the Statute Book up to Date: A Personal View’, CLB, Vol. 36, No. 79, 2010, p. 98.

• 11 Thornton, 1996, at 407.

• 12 C. Ilbert, Legislative Methods and Forms, Clarendon Press, Oxford, 1901, p. 259.

• 13 Thornton, 1996, p. 407.

• 14 Report from the Select Committee on Acts of Parliament (1875), p. iv.

• 15 Quoted in Berry, 2010, p. 81.

• 16 Lord Brightman, ‘Drafting Quagmires’, SLR, Vol. 23, No. 1, 2002, p. 1. See also Sutcliffe v. Commissioners of Inland Revenue, (1978) 14 TC, p. 171 and legislation in ‘unintelligible form’.

• 17 Northern Ireland Law Commission, Report on Bail in Criminal Proceedings (NILC 14, 2012).

• 18 F. Bennion, Bennion on Statutory Interpretation, 5th ed., Lexisnexis, London 2008, p. 605.

• 19 Lord Simon, ‘Statute Consolidation: Interim Techniques’, Public Law 1985, p. 352.

• 20 Lord Hailsham, ‘Addressing the Statute Law’, 4 SLR 1985, p. 7.

• 21 A. Samuels, ‘Consolidation: A Plea’, 26 SLR 2005, p. 56.

• 22 Justice (Northern Ireland) Act 2002, s. 51(1).

• 23 Law Commissions Act 1965, s. 3(1)(d).

• 24 Consolidation of Enactments (Procedure) Act 1949, Standing Orders of the House of Commons, Standing Order 58.

• 25 Guideline 18, Interinstitutional Agreement of 22 December 1998 on common guidelines for the quality of drafting of Community legislation (1999/C 73/01).

• 26 W. Robinson, ‘Manuals for Drafting European Union Legislation’, 4 Legisprudence 2010, p. 129.

• 27 Merkur Island Shipping Corporation v. Laughton, [1983] 2 AC 570, p. 570, at p. 595.

• 28 A.C. Lovgren, ‘The Location of Textual Amendments in Bills’, 9 SLR 1988, p. 27.

• 29 J. Beatson, ‘Common law, Statute Law and Constitutional Law’, SLR, Vol. 27, No. 1, 2006, p. 4.

• 30 Berry, 2010.

• 31 D. Hughes & H. Davies, ‘Accessible Bilingual Legislation for Wales’, 33 SLR 2012, p. 103.

• 32 J. Teasdale, ‘Statute Law Revision: Repeal, Consolidation or Something More?’ 11 Eur J L Reform 2009, p. 157.

• 33 Renton Committee, Preparation of Legislation (1975) Cmnd 6053.

• 34 (28 March 2001) 624 HL debs. col 264.

• 35 Brightman, 2002, p. 6.

• 36 Society of Public Teachers of Law, ‘Preparation of Legislation’, JSPTL, Vol. 13, No. 96, 1974, p. 99.

• 37 Para. 13.21.

• 38 Statute Law Society, Statute Law. The Key to Clarity (1972) para. 31. Statute Law Society, Statute Law: Radical Simplification (1974) paras101-120.

• 39 Samuels (1974, 1991, 1997, 2005).

• 40 T. Bates, ‘Monitoring of Current Legislation by the Statute Law Society’, 1 SLR 1981, p. 39.

• 41 C. Carr, ‘British Isles: Legislation’, 31 J Comp Legis & Int’l L 1949, p. 1.

• 42 M. Zander, The Law-Making Process, Cambridge University Press, Cambridge, 6th ed., 2004, p. 29.

• 43 Brightman, 2002, at 3.

• 44 Beatson, 2006, p. 4.

• 45 Hailsham, 1985, p. 8.

• 46 Renton, para. 13.21.

• 47 F. Bennion, Bennion on Statute Law, 3rd ed., Longman, London, 1990.

• 48 Bates, 1981, p. 39.

• 49 H. Xanthaki, ‘On Transferability of Legislative Solutions: The Functionality Test’, in C. Stefanou, H. Xanthaki (Eds.), Drafting Legislation: A Modern Approach, Ashgate, England, 2008.