!-- consent -->
The event: the annual Conservative Conference. The place: Birmingham. The year: 2012 – three years before the 2015 election in which David Cameron pulled off the biggest surprise win in modern times. This was the venue for the presentation by Stephen Gilbert, the Prime Minister’s Political Secretary, which outlined the strategy that – not without its hiccups and problems – was delivered on the ground and made victory possible.
Speaking in a closed session to senior activists, Gilbert set out the programme for the Tory Stealth Win that was painstakingly effected beneath the radar of Labour, the media – and indeed the pollsters.
What was it and how did it work? The Conservative air war was visible for anyone to follow – the assault on Labour’s fiscal credibility, the image of a weak Ed Miliband being propped up by a strong SNP, the starkly-drawn dividing lines on such as welfare reform and the deficit. These headline messages certainly played a large part in delivering the Conservative majority.
We are less familiar, though, with this other reason for the election result. The disproportionately good results in the marginals – both those the Tories needed to hold and those they needed to gain – are a testament to the success of the 40/40 strategy. The Liberal Democrats were devastated, despite their supposed incumbency advantage in the seats they held, with more of their voters moving to the blue column than most had expected. In many seats, it also seems the Conservatives were not only more effective at hanging onto their voters and winning over former Liberal Democrats, but were also more successful in wooing UKIP voters than Labour.
A good national campaign might have been a precondition for success in such seats, but a strong ground war was essential to delivering such good results. Three years ago, “the 40/40”, Team 2015, VoteSource and RoadTrip2015 weren’t even gleams in Cameron’s eye. Yet each came to play a crucial role in deciding the future of the country, defying almost everyone’s expectations and securing a historic result.
Today, we tell for the first time the story of that ground war – how the strategy came about, who controlled it, and how it worked in practice.
Lessons from 2010
Despite the glow of the Downing Street garden setting at that first press conference, the 2010 election was not an unalloyed success. Up against the worst Prime Minister for years, on the back of a financial and fiscal crisis and led by Cameron, whose poll ratings had long been more favourable than those of his party, the Conservative Party had failed to win a majority.
There had been serious problems with the national messaging – as Steve Hilton has recently acknowledged – but there were also problems with the ground campaign. After years of a decline in membership, and near-extinction in some areas where the Conservatives ought on paper to have been competitive, the party’s line regiments were variable at best.
Institutional knowledge, experience and enthusiasm were combined in some strong Associations, who continued to make do more with fewer people, but in other seats, even where there were people ready and willing to campaign, the tactics and techniques being used were either outdated or held back by the loss of experience, in the wake of the death or departure of older members. Some seats saw stellar swings based on a combination of good candidates and good campaigning – but others fell far short of even the national average Tory swing.
It took time for CCHQ’s gaze to move on to the next election. The initial throes of coalition had to be navigated, it lost many of its staff to Government as Special Advisers, and others departed for the private sector.
It’s also fair to say there was little sense of urgency to review what went wrong – after all, the Conservatives were back in government, albeit not with the majority which many had been led to hope for.
But by 2012 – at roughly the time of Sayeeda Warsi’s replacement as Party co-Chairman by Grant Shapps – CCHQ was once again looking forward. It was being comprehensively restaffed and, crucially, the “omnishambles” budget had occurred. It was a wake-up call for the Tory machine.
Conservative poll ratings had taken a nosedive and, for the first time, it began to seem plausible that Ed Miliband might be Prime Minister in 2015. The task of getting CCHQ into shape was beginning to look more challenging. Simultaneously, new tactics were being used in the US presidential election which sharpened the sense at the top of the Party that the Tories were lagging behind on the practice of political campaigning.
The seats – and targeting
It was in the summer and autumn of 2012 that the 40/40 strategy was born. When Gilbert briefed Party members on the plan in Birmingham, it sounded deceptively simple. The campaign would focus on defending 40 Conservative-held seats and attacking 40 others then held by Labour and the Liberal Democrats.
The seeds of the scheme lay in close analysis of the 2010 results. Even then, there were signs that good ground wars had defied the problems caused by the faltering air war. As Tim Montgomerie wrote at the time:
“The Conservative Party now feels it has a better ground operation than Labour. It points to the fact that it won 23 more seats at the last election than it should have done if the national swing towards the Tories had been evenly spread across the whole country.”
That factor saved the Conservative bacon in 2010: what if it could be extended and deployed, in addition to a more effective air war? Rather than simply stemming a red tide, Gilbert and others believed, it could produce a Tory majority.
By the time of Shapps’s appointment, the analysis and development of this idea was underway, and the new Chairman and the Prime Minister agreed with the campaign team’s conclusions. What they needed next was draw up a detailed process for identifying which constituencies were feasible prospects.
Defence seats were relatively simple: the Tory MPs who sat on slim majorities were obvious priorities, though it was made clear to them that support from the 40/40 process would require them to follow new approaches, and agree to do things they might instinctively question.
The more complex process was picking the attack seats.The Conservative top team rejected the traditional principle of basing targeting decisions purely on election results or overall majorities, and pursued data the team believed to be more meaningful – not just how many people voted Lib Dem in 2010, for example, but what sort of people are were. The demographic questions dug deep. Were they old or young, long-standing residents of the area or incomers, employed or unemployed? Did they own their own homes, hope to do so, rent privately or live in council housing? A multiplicity of data points were drawn up.
Not everyone on the ground agreed with the view from the centre. Seats which would have previously been viewed as natural targets were left off the list (such as Edgbaston in Birmingham or Barrow on the Cumbrian coast), while others with larger Labour or Lib Dem majorities were included. By 6th November 2012, the first ten attack seats were named and opened to applications from would-be candidates. Each would get a swift selection, and a professional campaign manager, to get them off to the best possible start.
The candidates – and “soft pressure”
During the 2010-2015, Cameron and CCHQ launched a full-frontal assault on the traditional image of Tory candidates – the A-list initiative. This overt drive to push candidates who were more ethnically diverse, more female and – sometimes bizarrely – more famous arguably caused more pain than gain, and certainly angered many activists at grassroots level. Resentment about the “parachuting” of candidates into safe seats was widespread.
In picking candidates to fight target seats in 2015 – and thus make up what the Conservative leadership hoped would be the new Conservative Commons intake – CCHQ faced a dilemma. It believed, not unreasonably, that a more diverse Parliamentary Party would not only be a gain in its own right, but that more women and more ethnic minority Tory MPs, in particular, would be a powerful aid to their attempts to win new votes. At the same time, they had just seen how seriously such an attempt could backfire if it was seen as a ploy from the top.
Instead, Shapps adopted an approach that his friends describe as “soft pressure”. The candidates’ list had been rebuilt, and was more diverse than before, but this wasn’t the only step. In some cases, particular potential candidates were invited to apply to be on the list before the selections for the 40/40 seats closed. Furthermore, the process for shortlisting was changed, so that the sift of CVs would normally take place at CCHQ – carried out by local Association officers, but with the advice of people from the central Party.
Ultimately, the decision still came down to local members – or local voters, where open caucus meetings were used – but they, too showed a willingness to select more widely than some stereotypes would suggest. As CCHQ had learned, and as the rise in female and ethnic minority Tory MPs last month shows, you catch more flies with honey than vinegar: persuasion had worked far better than the blunt force of the A list.
The voters – and deep research
Having picked the seats and selected the candidates, the next step was to understand key voters in each constituency. Lynton Crosby began working for the Conservatives in January 2012 (after calls from this site among others for the party to hire him), and part of his role was to lead on the polling and analysis of voters in each target constituency: how had they voted in the past, why had they done so, what might make them stick with the blues or switch, and so on. He believed there were more potential swing voters than people realised and, in particular, that Liberal Democrat supporters were more amenable to voting Tory than others realised.
This laid the foundations for the ground war: without it, the ground campaigning, phone calls and leaflets to come would have been far less effective. If the concept of the 40/40 strategy was a precision strike to win a majority, Crosby’s research (later bolstered by Jim Messina’s data) was aimed at delivering a precision strike to win a majority of votes in each seat.
The importance of getting the research right cannot be overstated, either in terms of the eventual election result, or the subsequent media confusion. The majority would be won by campaigns targeted directly at a relatively small number of groups, each composed of a relatively small number of people in a relatively small number of seats.
That level of detail was hard for those working at a national level to see, and Crosby’s own insistence that “we don’t talk about process” combined with it to make crucial parts of the campaign almost invisible. The Conservative approach was, in effect, a rather secret war, carried out below the radar of the watching national media, which had no means of assessing the quality of the information gathered in the databases, or all the ways in which it was duly exploited.
The troops – and their absence
But although the polling could identified potential voters, it couldn’t actually get to them – ring them up, send them e-mails, deliver them letters, knock on their doors. This is the point on which the election strategy turned. For the advances in social media, opinion research and campaign technologies, the blunt truth is that you still need boots on the ground to get the job done.
As I wrote last year on the need to rebuild the Conservative grassroots:
“Over a century and a half, the Conservative Party has built institutions on top of its grassroots base – philosophers, policy experts, technocrats, logistical specialists, copywriters, spinners, candidates, peers of the realm and so on. That’s all to the good…but when an organisation’s grassroots decline, those lingering functions of its centre serve to mask the scale of the problem. Campaigns are launched, pamphlets written, leaflets sent out, speeches made – but more and more it becomes a mere imitation of life. Eventually that underlying reality is exposed by a real test. An army, equipped with generals, vehicles, radio operators, cooks, artillery pieces and ammunition may look and sound very impressive as it rolls out to the front, but if its fighting troops, the tip of the spear, aren’t there then it will swiftly be found out in combat.”
The Eastleigh by-election campaign in 2013 was a shock to CCHQ. At its start, CCHQ was confident that the Tory ground operation would be strong – only to discover that the local party had been almost completely hollowed out. MPs were bussed in not to rally the troops but to be the troops.
With a developing 2015 strategy that relied on regular communication with target voters – voters who could only be identified in the first place by activists knocking on their door or ringing them up – there were fears that the party’s machine might not be up to the job.
At that time, the Conservative Party had experienced a decade of falling membership numbers. “To run an air war without a ground war would be like trying to run a car without gas or oil,” sums up one person who was closely involved in the campaign. Downing Street and CCHQ simply could not afford 2015 to become Eastleigh writ large.
The challenge in the by-election’s wake was two-fold: how to engage a wider circle of people than party members alone, and how to ensure they and the established membership could be deployed in the right places, the 40/40 seats.
Shapps himself came up with what would become at least a large part of the answer: Team2015.
Team2015 – “Reward and Recognition”
This new concept, conceived and agreed in the early months of 2013, announced at that year’s Spring Forum and the focus of a recruitment drive that autumn, imagined a new arm of the Conservative campaign machine. It would be free to join, open to members and non-members alike, focused entirely on the 40/40 and run by the centre in support of local campaigns.
The last point was key – Tory membership, like Tory votes, is “stacked up in the South, where we don’t need it”, in the words of one campaigner. It also piles up in safe seats, not in marginals, and there was frustration at the centre that activists tend to campaign where they live – often in seats that the Party already holds comfortably – rather than in the marginals where they might help to deliver a majority.
Messina, the former Obama campaign adviser hired in 2013, contributed input from the presidential race the previous year. “Reward and recognition” was key, he told other strategists, to attracting and retaining activists in such a programme. Elements of competition, as well as prizes for the hardest-working campaigners, were added – with input from Jeremy Hunt and others.
Still, all this was only a start. By November 2013 the project had 4,000 members – not enough to do the job which lay ahead. There was also a problem with the execution. “Team2015 was [at that stage] a dating agency,” remarks one insider, “it matched people up with target associations.” However, not every local campaign was then making full – or in some cases any – use of the new names. Other activists were travelling to their allotted seat once or twice – and then drifting away.
CCHQ still believed that the idea was sound, but that it needed to be grown in numbers and intensified in its push to motivate people. A number of factors combined to improve the process.
In January 2014, Shapps poached Paul Abbott from Robert Halfon’s office to be his new Chief of Staff. Experienced in Halfon’s successful model of grassroots campaigning in Harlow, Abbott was given the task of working out how to kickstart Team2015.
His assessment drew him to two conclusions: it needed more direct management, particularly of the process of getting activists out and campaigning, and it needed more money – rather a lot more money.
When Shapps presented his plan to his colleagues on the Executive Senior Management Team, he faced a tricky task. His co-chairman, Lord Feldman, had been instrumental in putting the party back onto a sound financial footing, and had developed a reputation for being careful with the Party’s cash. Team2015 would need around £300,000 during the 18 month run-up to polling day, a fair bit of it to be spent on buses, train tickets, hotel rooms and curries to get and keep people where they were needed.
Not everyone thought that this would be a good use of funds. Crosby was concerned that the proposed model of campaigning would prove “messy”, and potentially disruptive to the national message (indeed, during the campaign a couple of Team2015 activists were door-stepped by broadcasters trying to find out where they were from). Feldman, though, decided to back it and agreed the financial request.
From this point, the Team 2015 numbers grew drastically. Estimates vary, but by polling day the database contained somewhere between 75,000 and 100,000 people, only a minority of whom were party members. This is not to say that all of them were active. CCHQ’s records indicate that around 15,000 of them did at least one active piece of campaigning, though it’s likely that others turned up without the central system being informed (particularly during the last few weeks, the emphasis was on getting people out and damn the paperwork).
Getting people there – and co-opting RoadTrip 2015
During the course of 2014, the old “dating agency” model was scrapped and replaced with a more direct approach to managing the Team’s members. While everything was still co-ordinated with local associations (“you can’t have people turning up to find there’s no leaflets,” as one campaigner points out), CCHQ took control of booking in action days and getting people out to take part.
This “Super Saturday” model was about more than just appearances (though the photos of scores or hundreds of mostly young activists out on the campaign trail certainly perturbed the opposition). It was, in the view of those running the ground war, essential to ensuring that people actually came. While some activists were willing to spontaneously turn up to help in their allocated target seat, getting large numbers of them to target seats required a more structured effort.
CCHQ experimented with different approaches to motivate people, using campaigning software to test whether arranging to meet in a pub secured more RSVPs than arranging to meet on a particular street – and discovering, while doing so, the pulling power of a promise of a curry once the campaigning was complete. A “chivvying team” was established in Matthew Parker Street to use (seemingly endless) emails, phonecalls and text messages to chase up the volunteers for the forthcoming action days. Deborah Feldman played an vital part in co-ordinating the effort, and a running tally was kept of how many people had agreed to go where on a whiteboard above the phonebank.
CCHQ had also co-opted the RoadTrip2015 campaign model – which had been developed by Mark Clarke, the Parliamentary candidate in Tooting in 2010, during the previous year – and which delivered activists from elsewhere to campaign in target seats.
This proved difficult and costly to scale up, but buses and trains were provided to ensure that the right people arrived at the right place at the right time. Later in the campaign, six battle buses (real buses in the Midlands, the South West, two in the North West, a “SpAd bus” leaving from CCHQ each day and a metaphorical bus of 50 people heading out from London on the train) would be deployed to fill campaigning gaps in more remote or less well-staffed battleground seats. Most of the time, though, people were encouraged to make their own way.
The approach had its critics: a “Potemkin village”, some called it. This proved to be unfair. During the last 28 weeks of the campaign, Team 2015 supplied 26,000 campaigning days in the target seats. While this effort was not a replacement for the wider party (or for the contribution of other supporting groups, such as the pro-hunting Vote-OK group, which contributed campaigners in around 25 seats, or the various Conservative “Friends of” groups), it was an undeniably valuable contribution. If just 901 people in the most marginal seats had voted Labour instead of Conservative, last month’s majority would never have been achieved: every one of those days spent campaigning was crucial.
The campaign – and “The Survey”
But getting activists in the right place was only the first hurdle to jump. The next was deploying them correctly. Tom Waterhouse, the 2010 campaign manager in Enfield North, lays out the key stages of a campaign in a masterly blog here:
“Election campaigns have three phases: voter identification, motivation and GOTV (get out the vote). A good campaign will begin by using the first phase to identify the people minded to vote for you. In the second phase you spend your time and resources ensuring those supporters are sufficiently encouraged to turn out and vote for you (hence all the leaflets, letters, door-knocking and phone calls). Finally, you need the logistics in place so that GOTV runs smoothly and all your supporters know when polling day is.”
In the 40/40 seats, that meant kicking off with a tool discussed in Tory circles in a tone that implies capital letters – The Survey. With 14 questions, designed by Gilbert using the research from Messina and Crosby, this was intended to be a swift but effective way of identifying voters. The strategists worked to move away from a binary model of identification – Tory or not, Labour or not – and to collect more subtle information, rating people’s enthusiasm for various parties on a scale of 1 to 10.
Such data has a practical use: with the kind of hyper-targeted communications CCHQ was planning, it needed to know as much as possible about people’s interests and concerns in order to segment them accurately. As Waterhouse notes, if you don’t know where the relevant voters live by the start of the Short Campaign, you will have problems.
On the doorstep and on the phone, all those campaigning in a target seat were equipped with the same script, and sent out to gather the same data on a consistent basis. That said, there remains some difference of opinion even now as to whether The Survey was quite right – opinions expressed to ConHome vary from “fantastic” to “way too long”.
Delivering the right material to the right places depended on slick co-ordination between the air war and the ground war. Numerous candidates and MPs – before and after polling day – have told me that the link between the two worked far better than in the past. Crosby and Gilbert crafted messages carefully tailored for at each target voter, and the activists ensured that these were delivered. Local issues and personalities were built in to the material – through close work with the candidate, campaign manager and Association in each area. And there was a strong emphasis on ensuring the national messages were delivered locally.
For example, on the Saturday before polling day I was campaigning in Sutton and Cheam, my own nearest 40/40 seat. We were delivering hand-addressed envelopes to voters known to be wavering Liberal Democrats. I took the opportunity to open a few of the envelopes, and was impressed to find a variety of different messages inside. Some addressed the dangers of a Miliband-SNP government, some talked about the jobs miracle, others about the deficit. The campaign team there had a good sense of who their voters were, what motivated them and which ones might be willing to change their minds – invaluable knowledge on all counts.
The computers – and the crash
However, there was one dark cloud on the campaign’s horizon. Having carried out the voter identification and completed the motivation phase, all that remained was to Get Out The Vote. All those tens of thousands of days surveying people, leafleting and door-knocking led inexorably to polling day itself. Over the final days of the campaign, the regularity and tone of the exhortations to the grassroots to turn up and help on the big day itself intensified: those involved attest to a mounting sense of urgency.
The battle buses increased in number, to ensure as many people were on the ground as possible. Expeditions by train were planned. “The chivvying team” redoubled their messages and phonecalls to get as many attendees to the target seats as possible.
Then the computer system broke.
It didn’t break everywhere, or all at once. But in a troublingly large number of target seats, those staffing the campaign offices found that they couldn’t print out the slips to hand to waiting activists so that they could go out and “knock up” the identified Tory voters. The system was clogged up with all the data requests, and proved unable to serve all the constituencies at the most crucial time. Some local campaigns were paralysed, while others were able to turn to the hard copy canvassing returns they had kept in boxes, or at the back of cupboards.
The logjam was eventually cleared, and the Conservatives secured their majority despite it – but it was a heart-stopping moment. Its roots lay much earlier in the planning stages of the campaign.
When he took over from Sayeeda Warsi as Party Chairman, Shapps found that Merlin, the old IT system, needed to be replaced. It was rickety at best, took ages to input new data such as updates to the electoral roll, and was prone to losing information (memorably, it ditched at least one region’s entire set of canvassing data in the run-up to the 2014 European elections).
Furthermore, it wasn’t cut out for the kind of campaign the Conservative strategy required for 2015 – since it was unable to collect the more nuanced information that the new targeting required, to absorb the MOSAIC socio-economic database, to link up with new campaign tools or to allow the running of detailed analytics.
The grassroots disliked it, the campaigners couldn’t rely on it and the strategists needed something better: Merlin had to go.
A debate then followed about how best to replace it. Should CCHQ buy off the shelf – getting a product like NationBuilder, which is widely used in the US and formed the basis for Labour, Lib Dem and UKIP campaigns here – or develop their own system in-house, to precisely fit CCHQ’s needs? The fateful decision was made to go with the latter option. Shapps delegated the running of the project to Gilbert, who placed Paul Bolton, the Chief Information Officer, in charge. Bolton assembled a team and set them to work on what would become VoteSource.
“It was madness,” remarks one individual acquainted with the process. A specialist in the area of campaign technology describes the decision as “reinventing the wheel”, arguing that CCHQ could simply have bought a ready-made product which had been tested in the heat of battle elsewhere.
The sheer expense aside, the timescale was pressing: the huge task of building the system meant that it wasn’t ready to be introduced until early 2015, a short time before the campaign began in earnest. Effectively, the party had to run both it and Merlin side by side; the rush to introduce the new system meant it underwent various updates and fixes during the course of the campaign, and activists were being trained to use it right up until the last week of the campaign.
The meltdown on polling day was not the first of VoteSource’s problems, either. Numerous canvassing and survey returns from action days couldn’t be inputted, and were left in a cupboard in CCHQ – valuable data which could not be exploited.
It also failed to connect reliably with Connect2015, the new phone canvassing system. Both problems undermined what was meant to be the purpose of developing VoteSource in the first place – swift, reliable compilation of the data gathered by the Conservative grassroots.
Tellingly, not all of the party’s campaign operation ended up using the system. Team2015 itself, for example, used NationBuilder to manage its growing activist database, apparently without any technical hitches.
VoteSource undoubtedly has potential – and, whether in the form of an improved version of this system or an off-the-shelf alternative, the party clearly needs a modern data management system on which to run its future campaigns. But the serious consequences of that decision to build such software from scratch – the cost, the delay and the eventual disruption to the campaign – cannot be ignored.
Other aspects of the technology worked far better. Connect2015, which allows activists to make campaign calls to voters from home rather than trooping in to a campaign office, worked well even if VoteSource would not automatically incorporate its data. The fruits of Messina’s advice on encouraging competitive spirit can be seen in its leaderboard, where activists rack up points for successful calls. Some Associations certainly welcome a new way to involve elderly activists in their campaigning more easily.
The digital team at CCHQ – led by Craig Elder and Tom Edmonds – were key to the election win, through their use of social media to identify supporters, deploy key messages and raise money. Fundraising online, via small donations from large numbers of people, grew drastically during the course of the campaign, and is viewed by Conservative fundraisers as an essential growth area for the next few years.
A danger comes with winning elections. Just as when you lose you can end up thinking that everything you did was wrong, victory can lead you to believe that everything you did was right. Better to make an open and honest assessment of what worked well and what didn’t – and of what needs to change to ensure future victories.
Team2015 was a success in its own terms. Rather than providing a mere media facade of active grassroots, as some claimed, it delivered a sizeable amount of valuable campaigning resource to target seats. Its designers and controllers experimented and learned as they went along, ditching experiments that didn’t work and improving the model. But it is conceded, even at the highest levels in CCHQ, that the project only just got the Conservatives over the winning line – and that it is essential not to have to start from scratch again in 2017.
Nor is Team 2015 a permanent answer to the challenges the Conservative Party faces. Without a reliable Association structure to hold the line in other seats, targeting and focusing resources is like icing without a cake on which to sit. Without a truly mass membership, the Party could be defeated by a more capable, forewarned and better-organised Labour Party next time round. Lord Feldman’s review of the Party’s structures and membership model should probe the lessons of what worked this time – but also look to the future for new ideas.
The least successful aspect of the campaign was the decision to commission VoteSource, apparently without sufficient time to get it right. The Party has long been plagued by poor IT, and that it won through, despite these problems, was due to three factors: the strong performance in the air war, the hard work of activists in the “motivation” stage running up to polling day and the well-planned and well-executed process of voter identification beforehand. It must now identify the VoteSource problems and fix them or – if they are too great for that, get something new off the shelf.
The 40/40 strategy, and its attendant tactic of hyper-targeting voters based on detailed information, was clearly effective. If in 2020 the “defence” seats are those gained in 2015, and the attack seats are a whole new tranche of Labour territory to eat into, the next stage of the contest on the ground will be intriguing.
However, there are at least two sides to every electoral contest. Labour and others will be studying the 2015 results closely, in an attempt to correct their mistakes. Ultimately, winning an election is a short-term enterprise. One simply moves on to the next one – in this case, 2020, with 2025 beyond – and to the challenge of doing better among groups who tend not to vote Tory: younger, urban, liberal, ethnic minority and public sector voters.
The final lesson is about the combination of skills and judgement required for a successful campaign. Without Shapps’ willingness to recognise the scale of the problem, Gilbert’s commitment to message discipline, Crosby’s crafted messages and Elder and Edmonds’ social media nous, victory would not have been won. But perhaps the fate of the whole ground war can be traced back to a single decision: the moment when, early in 2014, Feldman was asked to back Team 2015 with a sizeable budget. He did so. It was the right judgement call. The rest is history – and lessons for the future. Had he ruled otherwise, ConHome would be writing a very different story.