Title: The Uncompleted Centennial Atlas

Author: Malcolm McKinnon

In: Creating a National Spirit: Celebrating New Zealand's Centennial

Publication details: Victoria University Press, 2004, Wellington

Part of: The New Zealand Centennial Publications

Keywords:

• Literature
• New Zealand History

Conditions of use

11: The Uncompleted Centennial Atlas

Malcolm McKinnon

The decision to produce an historical atlas as part of the centennial celebrations for 1940 was taken by the National Historical Committee at its inaugural meeting in June 1937. An atlas and gazetteer subcommittee was appointed under the chairmanship of a member of the committee, Professor James Rutherford of Auckland.¹ The atlas would contain 'both an historical and a modern section.'² That atlas was never completed, and has inevitably tended to become invisible in discussion of the centennial, yet at the time it was widely expected to become a centrepiece of the publishing programme. As Heenan put it in a submission to cabinet, the atlas would, he was satisfied

be the outstanding, and, next to the Dictionary of National Biography, probably the most important of all the Surveys. So important does the National Historical Committee regard the Atlas that it recommended an enquiry into the possibility of its being published jointly by the Government and the Oxford University Press. ... an endorsement by that press of the work as something outstanding, and so, incidentally, secure practically a world market.³

Even taking into account the hyperbole of a public servant wishing to reassure a minister (and beyond the minister a cabinet) of the merits of a course of action, there was some justification for this view. The atlas was to be both big and bright, with a 16¹/₂" by 13¹/₂" map plate size, later altered to 17¹/₂ by 11¹/₂ (430mm x 290mm)— presumably because it was realised that New Zealand was long rather than wide! This was larger than the dimensions of the Historical Atlas published in 1997. The proposed atlas entailed a lavish use of colour (up to eight colours for one map series— in contrast the 1997 atlas used only the four process colours) and the individual sewing of the map pages into the binding. A total of eighty-eight of these map pages, comprising eighty-five separate maps (some would spread across the middle) was envisaged. Total costs, including drafting, printing and binding, for a print run of 3000, were estimated at approximately £10,000, or a unit cost of around £3 6s.⁴ In sum, the atlas was to have been a monumental enterprise. The surveyor general's report was approved by the atlas subcommittee at this time. It was expected the atlas

¹ Ms-Papers 0230-239, report to National Historical Committee (NHC), ca June 1938.

² Appendix to the Journal of the House of Representatives (AJHR), 1938, H22, p.3.

³ Ms-Papers 0230-239, Alexander Turnbull Library (ATL). Extract from memo, Undersecretary to Minister of Internal Affairs, 5 July 1938.

¹ Ms-Papers 0230-239, report to National Historical Committee (NHC), ca June 1938.

¹ Ms-Papers 0230-239, report to National Historical Committee (NHC), ca June 1938.

¹ Ms-Papers 0230-239, report to National Historical Committee (NHC), ca June 1938.

¹ Ms-Papers 0230-239, report to National Historical Committee (NHC), ca June 1938.

⁴ Ms-Papers 0230-001, ATL. Surveyor General to Secretary, NHC, 7 September 1937, 13 September 1937.

page 150

Maori and European land ownership in the North Island, 1860. Alexander Turnbull Library, Wellington, NZ, MapColl CHA-6/2/3

would sell at cost; but at a time when £3 6s was several days' pay for many wage-earners, this would still, it was recognised, put it out of the reach of most ordinary people.

Such an ambitious project naturally raises questions. Why was it taken on? And why did it fail to appear? The records are not illuminating on the first question, but the general mood of the centennial celebration project, in which so many different schemes were canvassed, must have played a part.

Rutherford, of the history department of Auckland University College, and H.E. (Harry) Walshe, surveyor general and head of the Department of Lands and Survey, were the two people most prominent in the early deliberations about the atlas. Rutherford was a historian of British colonialism. He had done work in South Africa as well as New Zealand (his magnum opus was to be a biography of George Grey, who served in both countries) and was wont to draw sketch maps himself.⁵ He was not a particularly collegial individual, and his distance from the project was to be reinforced, firstly, by his being resident in Auckland, and secondly, by his taking sabbatical leave away from New Zealand from August 1937 until the middle of June 1938.

Harry Walshe was the Crown's surveyor general, and chief of the Department of Lands and Survey. A quiet 'lovely' man, his first great enthusiasm was map projections, his second, fishing in the Marlborough Sounds.⁶ As a government department with a lengthy tradition of map-making expertise, it was inevitable that Lands and Survey be involved in the preparation of an atlas, and Walshe was asked to join the atlas subcommittee from the beginning. It is clear that Walshe saw his department as the 'producer' of the atlas and he was prepared to stump up at least half of the cost from his departmental vote. If the National Centennial Committee met the cost of binding up to £5000, Lands and Survey would meet all other costs, he said. In a second memo, provision for binding was settled at £3000/

The atlas subcommittee in effect became the manager of the project. Rutherford and Walshe appeared to have reached agreement on the choice of maps, which included both historical and thematic or contemporary maps, plus coverage of contemporary New Zealand in a number of plates at eight to ten miles to the inch (about 1:500,000).⁸ The subcommittee continued to meet while Rutherford was on sabbatical leave, chaired by Walshe and serviced by E. H. McCormick, as secretary of the National Historical Committee. In March 1938 cabinet approved the atlas as a charge against Internal Affairs' centennial vote.⁹ At this juncture Heenan, who had overall responsibility for the publications programme, added both himself and Oliver Duff, the editor of the centennial surveys, to the atlas subcommittee. But Lands and Survey was to bear one third of the cost of the atlas, covering that part comprising the large-scale area maps 'being prepared by Lands and Survey as part of its ordinary work' (that is, the eight or ten mile to the inch series).¹⁰ This was in fact the only centennial publication project which involved an interdepartmental collaboration.

War broke out in September 1939. In a letter of 12 October Burnett wrote that the date for printing the atlas was still under consideration, but that it seemed 'fairly certain that as a result of the war, this may be postponed for some little time. But the rest of the work, such as research and drawing, is continuing as before'.¹⁵

Within a very short time, however, this proved overoptimistic.The sharp rise in the cost of paper after the outbreak of war was one problematic factor: the cost of printing was estimated at £2849 (for a run of 6000) in September 1939 and £5698 in February 1940.¹⁶ But more immediate in its effect was the withdrawal of draughtsmen from the project. When the atlas had first been authorised Lands and Survey had dedicated a team recruited from all its district offices to the project. But three of these enlisted for overseas service, another took on territorial training and a fifth was transferred to another department. The other cartographers were faced with such a big programme of defence mapping that they too were effectively lost from the project.¹⁷ The final drawings were 'put aside' in July 1940. There was a prospect then that the atlas would never appear.

If this is the bare bones of the story of the project, 1937 to 1940, to where do we look for an explanation of the failure? In effect there is a hierarchy of explanations, the most immediate being the war, as discussed above; among less direct influences, three intertwined considerations are crucial: production, management and content.

Who was to do this work? The text was to be the responsibility of scholars, the gazetteer could be produced by the mapmakers. But the maps were a different matter. If this had been a contemporary atlas there is no doubt that the management of it would have best rested with Lands and Survey, which was 'undertaking the research for a number of the maps to be included in the Atlas'. That department knew how to compile contemporary maps, be they general or thematic (for instance of forest cover).

Harbours, shipping routes, lighthouses and radio beacons around Otago and Southland. Alexander Turnbull Library, Wellington, NZ, MapColl CHA-10/3/20

It either collated the necessary information itself or gathered it from the relevant government departments. Many of the thematic maps envisaged for the atlas were prepared in this fashion.¹⁹

With respect to the historical maps, however, no such practice could be followed. Such maps required compilers with expertise in historical research, who had then to work with the cartographers. This was not straightforward. The task of making the information gathered mappable required a familiarity with the mapmaker's art. Whereas for the researcher, the information was the map, for the mapmaker, the legend, or key, was the map. That was what provided the mapmaker with the 'driving instructions' as it were. And not only did there have to be such instructions, they had to be readable, or rather, allow the production of a map which was readable. In other words, there was a challenge in converting one language—that of historical research—into another, a usable map.

This problem might have been overcome if the management of the project had been more robust. Who did manage it? The short answer is—no single person. A slightly longer answer is the atlas subcommittee. There are enough aphorisms about management by committee, but no easy answer to the question of why the committee did not appoint a manager or editor.

One possible answer lies in a version of the contrast between historical research and mapmaking already touched upon, as there is some evidence that this reverberated institutionally. The uncertainty about what the atlas should be called is suggestive, the two commonest appellations being 'historical atlas' and 'centennial atlas'. Sometimes both adjectives were used, quite often neither. McCormick in August 1937 referred to the 'historical atlas', but Walshe in September to the 'centennial atlas', as did historian W. H. Skinner some months later.²⁰

Beyond such matters the need for two departments to collaborate also produced some predictable tensions. One occurred around the time of Rutherford's return from leave, and resumption of the subcommittee chair. Sixty years and more later it has a slightly comic aspect which may not, however, have been evident to the participants. The atlas was to include a number of facsimile maps, such as Tasman and Cook's charts and the 'Tuki' map, a map of New Zealand drawn from information supplied by two Maori informants to British officers on Norfolk Island in 1791. The subcommittee learnt that the draughtsmen proposed to 'redraw' these maps in some respects, partly to increase the legibility of their texts. This seems to have been accepted, if not very enthusiastically ('many contentious points had to be decided' record the minutes of a meeting of 11 July) by other members of the subcommittee when the proposal was explained by Walshe and his deputy R. G. Dick. Ironically, the Lands and Survey official who had carried out the redrawing, W.G. (Guy) Harding, was the father-in-law of J. D. Pascoe who was employed by Internal Affairs, partly on work for the atlas, at the time, and was present at the meetings on this issue.²¹

The historians were not happy. Rutherford was to refer to the 'grave objection' there was to redrawing at the subcommittee meeting of 19 August (the first he had attended since returning to New Zealand), and Dr J.C. Beaglehole, a member of the National Historical Committee who had a keen interest in Pacific exploration, also became involved. An 'informal' meeting at Internal Affairs at the beginning of August, which Beaglehole attended, concluded that 'the difficulties which had arisen could only be prevented if the position of the Lands and Survey Department as the agent of the National Historical Committee and of the Department of Internal Affairs in the production of the Atlas was clearly laid down'.²² At the 19 August meeting this plainly occurred, as the decision to redraw three maps was overturned, with the two Lands and Survey members dissenting. They must have recognised the political realities of the committee, however, as the minutes record the committee agreeing, without any dissent, that all drafts should be approved by the atlas subcommittee.²³

Did this episode make Rutherford distrustful of Walshe and his staff? He examined the proposed atlas contents as set out by Walshe in a letter of September 1937 (after Rutherford had gone on leave) and decided that it had entailed a dramatic dilution of the historical content of his own original list, with which he thought Walshe had concurred. As he wrote to McCormick in an angry five- page letter penned on Empire Hotel, Wellington, letterhead, he now realised that many of his historical maps had been amalgamated into one, whilst it was intended that the nineteenth-century war maps would be overdrawn on the ten miles to the inch general maps and that the world maps would appear at the beginnning of the atlas. This would not do. 'We are concerned primarily to produce an Historical Atlas. The approach is historical, and, as far as possible, we begin at the beginning, and let the story of NZ's development unfold itself gradually in a series of maps, beginning with Maori explorers and Pacific navigators and coming up to the present day.'²⁴

What happened after Rutherford's outburst? The immediate aftermath of it is unclear, although it is recorded that Rutherford's memo was discussed at a meeting of the atlas subcommittee at which he was not present, and at which Walshe was in the chair (presumably Rutherford had by this time returned to Auckland),²⁵ after which further discussion was to be held with Rutherford. The next meeting, which was chaired by Rutherford, took place in December, and the absence of comment about Rutherford's concerns suggest that they had been allayed or had waned. Attention instead was directed to a proposal to expand the content of the atlas.

In November the project had been visited by K. B. Cumberland, then assisting Professor George Jobberns in the newly established geography department at Canterbury University College. There was evidently considerable examination of an Atlas of Finland and discussion of the need for more social and economic content in the atlas.²⁶ In a lengthy memorandum to Heenan, submitted just after this visit, Burnett set out the case for a bigger atlas, in this instance citing the Atlas of the Historical Geography of the United States as a comparable instance.²⁷ Burnett,

²² Ms-Papers 0230-239, ATL. Minutes of meeting of Atlas subcommittee, 11 July 1938; record of informal meeting, 5 August 1938.

²³ Ms-Papers 0230-239, ATL. Minutes of meeting of Atlas subcommittee, 19 August 1938.

²⁴ Ms-Papers 0230-239, ATL. Letter of 20 August 1938.

²⁵ Ms-Papers 0230-239, ATL. Minutes of meeting of 20 September 1938.

²⁶ Ms-Papers 0230-239, ATL. Department of Lands and Survey to Secretary, NHC, 21 November 1938.

²⁷ Ms-Papers 0230-239, ATL. Memo, Burnett to Heenan, 23 November 1938.

page 156

Urban land use in Takapuna and Devonport about 1940. Alexander Turnbull Library, Wellington, NZ, MapColl CHA-10/2/5

Beaglehole and McCormick had worked through that publication identifying themes that might profitably be covered in a New Zealand atlas, such as military and other special settlements, political parties, foreign elements in the population, ports and trade, and reproduction.

The approved extra content amounted to another thirty maps or fifteen half-pages.²⁸ At a meeting on 1 December the decision was taken to expand the content along the lines proposed, with Heenan and Rutherford adding the caveat that it remained important that the atlas be published in 1940.²⁹ More staff were hired, first R.R. (Rex) Cunninghame, later O.S. (Sam) Meads and W.S. (Mick) Borrie, two young historians. Meads and Borrie left for England in the middle of the year but were replaced by F. Lingard. Cunninghame later became a diplomat, Meads a high school principal, and Borrie an academic in Australia.³⁰ In the new year, young Davidson, now in Cambridge, wrote to McCormick, congratulating the team on 'the general direction of thought regarding the Atlas and the proposed expansion of its scope' both of which he thought 'splendid'.³¹

Maori tracks and waterways arund Waikaremoana. Alexander Turnbull Library, Wellington, NZ, MapColl CHA-3/2/11

And a further significant expansion of content took place as a result of an initiative by Apirana Ngata. An exchange between Heenan and Ngata took place in September 1938.³⁵ Renewing his interest in the project a year later, Ngata was particularly impressed with Cunninghame's work on place names and Burnett's general approach to Maori content.³⁶ Both aspects presumably allayed some of the concerns he expressed in Parliament a year earlier, for instance about identifying fortified pa associated with northern massacres of southern tribes.³⁷

Ngata proposed a series of maps of different parts of the country, with common waka affiliations, showing place names and other information connnected with that waka tradition (the scheme provided a model for the Tapatuanuku' plates of the 1997 Historical Atlas). After meetings with Ngata in September 1939 Burnett advocated a much more complete coverage of tribal history and place names. This proposal was also duly adopted by the atlas committee (the 'sub' had been dropped in April 1939).³⁸ In a surge of endeavour in 1940, a number of Maori were involved by Ngata in the project, amongst them Tupito Maruera from Taranaki, Tai Mitchell of Te Arawa, who had strong Labour Party links, Sam Maioha from Te Tai Tokerau, and Pei Te Hurinui Jones, of Ngati Maniapoto. Ngata for a time had a work space in the project's office, and both he and Jones had keys to it.³⁹

So, while the draughtsmen may not have understood the requirements of history, the historians did not fully grasp the requirements of a map production schedule. Either way, the centennial atlas was not to be. The tabulation of maps prepared when the final drawings were 'put aside' in mid-1940 suggests that perhaps a third of the atlas was ready, rather more than a third had not been started, and somewhat less than a third was work in progress. The completed third included a number of facsimile reproductions of old maps and Lands and Survey's special favourites, the ten mile to the inch series maps—on balance, therefore, more than half of the original cartography planned for the atlas remained to be done.⁴⁰

The goal of producing an atlas as part of the publications project for the centennial was a commendable one. For one thing, it could have fostered a greater Pakeha comprehension of Maori history, both pre- and post-contact—'a picture is worth a thousand words'. But the project was far too ambitious a one, even if all production and management issues had been resolved and the war had not supervened. As it was, it produced many impressive draft maps, both in this initial phase before the war, and as a result of continued work on it during and after the war. Research and map drawing continued until the project was closed down in 1951, but at no stage were the resources allocated to it adequate to allow it to be brought to publication.⁴¹ Many of those maps were of great assistance in the preparation of the New Zealand Historical Atlas more than fifty years later. And many remain valuable research aids to scholars to this day.