Faking global health

ABSTRACT

Globally, human health is improving. Aggregate world health data indicate enormous improvement over the last 100 years. Life expectancy, vaccination, and sanitation rates are higher. Rates of infectious disease, HIV/AIDS, child and maternal mortality are lower. These gains have all been accomplished during a time when governments orchestrated, or aspired to provide, albeit often imperfectly, the systemizations of health care. Now austerity and privatization campaigns shape health services worldwide, and we witness a massive ideological shift in approaches to the world’s wickedest health problems: Public health endeavors must ‘show return on investment.’ This commentary takes up activities in three health domains where effort goes into the appearance of global health prowess and accomplishment: health security; health innovation; and health finance. Pseudo global health, as an analytic, helps us take measure of the in-between phenomenon between real and fake accomplishment, success and failure, improved health outcomes and continued suffering. I show: 1) how global health security documents sometimes serve as ‘alibis’ – that is, contrivances offered to intimate local readiness or safety despite their actual absence; 2) how global health innovation influencers often privilege tech-fixes developed far removed from real-time people, places, and practices; and 3) how global health finance has already evolved in ways that makes suffering profitable. The examples are meant to enlighten rather than depress and are offered with the hope that critical analyses using the pseudo health concept as an analytic prompt new strategies for sustainable changes rather than merely their appearance.

KEYWORDS:

• Global health security
• global health innovation
• global health finance

Globally, human health is improving. Statistical evidence from the last 100 years shows uneven gains and backslides in some locales, but aggregate world health data indicate enormous improvements (Rosling 2010). Overall rates of life expectancy, vaccination, and sanitation are higher (GBD 2018: OECD 2019a; 2019b; IHME 2018, respectively). Average rates of infectious disease and child and maternal mortality are lower (IHME 2018). In recent decades, HIV/AIDS has been transformed from a fatal disease to a chronic condition when regularly treated (Ford et al., 2018). Even when we include what anthropological research reveals about health data and its fabrications (e.g. Biruk, 2018; Erikson, 2012; Kingori & Gerrets, 2016), there is no denying that statistical averages show impressive gains. Glassman and Temin deem it a ‘global health revolution’, one ‘that keeps mothers and babies alive, helps children grow, and enables adults to thrive through and beyond their working lives’ (2016, p. 1).

This commentary focuses on the work still to be done in global health. (Global health in this usage is the social, political, and economic fields shaping health praxes the world over.) Admittedly, the title, ‘faking global health’, is meant as a provocation. More precisely, I illuminate global health spaces and temporalities where common good has been assumed but in which actual accomplishment is blurrier, messier, and much less certain. Using pseudo global health as an analytic (see Kingori & Gerrets, 2019), I take up the activities of both local and expatriate global health actors to reveal those slippery social spaces between actual life-saving and appearances of it. Pseudo global health is a new way of eliciting unvarnished actualities within continuums of care and neglect, between real and exaggerated claims of accomplishment, so that we can take fuller measure of what constitutes truly improved health and wellness outcomes.

To make these points, I draw on previous ethnographically-based analyses (Erikson, 2017, 2018) and on-going ethnographic research in Sierra Leone and world cities where global health decisions are made. The focus is on three global health domains where effort has gone into the appearance of global health accomplishment: health security; health innovation; and health finance. I aim to problematize the sales-speak of their global public health promotional campaigns, which are performative in different ways. For example, health security in Sierra Leone prior to the 2014 Ebola outbreak did exist – on paper. The Ministry of Health and Sanitation, along with supporting multilateral and non-governmental organizations (NGOs), complied with International Health Regulations (IHR) paperwork requirements (now also digital). These requirements assure the world that health surveillance systems are in place to detect disease threats. The appearance of disease preparedness was good enough for the international community, until it wasn’t, as in 2014 when Ebola broke out. And in the burgeoning domain of health innovation, there is momentum and excitement about cultivating fresh new ways of developing health products and technologies (WHO 2019). Universities now house global health ‘incubators’, that is, institutes, centers, and laboratories designed to produce newfound remedies ripe for investor financing. But some of this innovation is more hype than help, as it was during the 2014–2016 Ebola outbreak when computational epidemiologists heralded big data technologies allegedly capable of detecting and tracking people sick with Ebola. The appearance of contagion-control innovation was enough to justify the attempted cooptation of West African telecommunications companies’ time, data, and attention during a pandemic. Computational epidemiologists insisted their containment model would work, until it didn’t. And, finally, in global health finance, we are in the throes of massive experimentation in humanitarian aid and global health funding. Impact bonds, blended finance, angel investing, and finance facilities are just a few of the new categories of ‘the third wave of aid privatization’ (Tricarico & Sol, 2016, p. 54). There has been a move from government provision and orchestration of essential health care and emergency response to Wall Street financing (used here metonymically to mean all capital markets). In global health, financial innovation has come to mean the creating and popularising of capital market instruments, like new insurance and pandemic bonds. New financial ‘facilities’, for example, appear to be easy remedies to the increasing reticence of wealthy countries to give money to poor countries for emergency response or assist in the cultivation of health systems. These new financial forms are well into establishing themselves as permanent fixtures in global health.

‘Modern’ global public health methods require designations of space, time and resources that are not innocent. Contemporary approaches build on old hierarchies that protect some people while compromising others (e.g. Hoffman, 2016). As I show, some global health security exercises are mere cut-and-paste exercises. Some global health technologies lead to no real sustainable health gain or improved outcome. And some new ways of financing global health response are converting health data originally intended for public health surveillance and accountability into data for invest-ability.

Global health security and the problem of document alibis

From March 2014 Fieldnotes (edited): About one week after two people in eastern Sierra Leone were identified as possibly sick with Ebola, a ministry organizes a meeting in Freetown, the capital city. Its purpose is to develop an Ebola Emergency Strategic Plan. About 30 government and NGO representatives attend. An administrator begins the meeting with an overview of Ebola and the threat of its spread. Then he moves to the task: ‘We have a template, but we need to bring it home, to make it Sierra Leonean.’ He explains that the template is ‘thanks to Uganda,’ meaning that the template is a World Health Organization (WHO) document from Uganda that needs to be wordsmithed for Sierra Leone. ‘We are here to make surveillance and laboratory plans,’ he said.

People in the audience respond as though they’ve done this before. Many of the people in the room had worked on cholera preparedness documents and operated as if some of that language could carry over to Ebola. The group begins discussing surveillance tools – reviewing the standards for evaluating suspected and confirmed Ebola cases; scouring the investigation forms ministry staff were expected to fill out with each suspected case; and reading through outbreak criteria lists.

The meeting participants begin wordsmithing the surveillance, laboratory, and treatment standards. The group lists various surveillance tools: a field guide booklet, step-by-step charts, blood biopsy kits, an infection control guide. But the conversation goes off topic before the list is completed. The group is distracted by talk about the budget. People begin debating about the number of people that need to be trained for RRTs (Rapid Response Teams). People calculate that with 1200 Public Health Units (PHUs) (health posts) throughout the country plus private sector clinics, 2 RRTs per PHU means that 2500 people need to be trained.

Then there is a brief kerfuffle about the number of triple sealed specimen containers needed to cover all the PHUs, and one administrator declared the country needed at least 500. One man asks, ‘Where did you get 500?’ She said, ‘I say 500 so that we can build capacity!’ He asks, ‘You just like the number?’ She said, ‘No, I like 1000, but I divided it by 2!’ People laugh at the arbitrariness of the calculation. Someone else makes the point these specimen kits will likely expire before any subsequent epidemic, and that capacity building won’t be met through over-requisition. The group settles on 100. Later, they field the idea of bringing Ebola victims to a Freetown hospital for definitive care. Someone points out that radically contagious patients shouldn’t be brought to urban centers. Another points out that the global treatment standards are ‘premised on the idea that public health systems already exist.’

In Sierra Leone, and elsewhere, paperwork – both actual paper and digital forms – often ‘fulfill’ international regulatory obligations and schematics. But these forms ‘paper over’ the ways in which a robust-enough primary healthcare system has not yet been made. The failure of attempts to build a healthcare system is not Sierra Leone’s fault alone. Wealthy countries have generally failed to acknowledge the ways World Trade Organization and international financial institutions (IFIs), like the World Bank and the International Monetary Fund hobble the implementation of a health system in Sierra Leone (e.g. Kentikelinis et al. 2015; Zack-Williams, 2012).

In the meantime, global health security documents stand-in for preparedness, providing a document alibi – that is, contrivances that are offered up to intimate local readiness and safety. Throughout my research, as I moved from office to office talking with people about health programs, I found that documents and readiness checklists stood in for actual preparedness. When conversations turned to health security and preparedness, 8 out of 10 interlocutors would pull a binder off the shelf or pull a document file up on the computer to show readiness. This ersatz document alibi-making was not a product of malice or deception but rather is the norm of the contemporary ad hoc non-system of health in many poor countries around the world: having the document demonstrates a willingness to be prepared, if not always actual preparedness. Paperwork is the work documents do in global health networks that require the appearance of viability.

The heralded IHR (2016) are another example of how, despite good international intentions, the bureaucratic energies devoted to making and maintaining the document and its regulations are over-done and disproportionate. After the 2014–2016 Ebola pandemic, so much time and expertise in the WHO and global health security community was spent on reviewing the IHR (e.g. Gostin, DeBartolo, & Friedman, 2015; WHO 2015). Compare that to the dearth of international political moxie needed to support the building of a Sierra Leonean national healthcare that would be a formidable first line of infectious disease defense. Herein lies the problem: In many impoverished countries, the contemporary historical moment is one in which health regulations – literally, documents themselves – sometimes stand in as response-ready mechanisms. The more difficult, necessary and fundamental task of health security is the building of quotidian clinical response capable of thwarting disease in the first instance. Infrastructural-intensive healthcare services are the actual sine qua non.

Case in point: In the 70-odd years since the inception of the WHO and after almost 60 years of Sierra Leone’s independence from Britain, the hard complex work of building Sierra Leone’s primary line of care remains largely undone. Distractions from this task have been multiple. The NGO-ification of the health sector means that each donor imposes its own set of templates, standards, and accountability documents, which local and expatriate NGOs spend a disproportionate amount of time preparing. Documents are understood as necessary proof of global health work, even when that work has not been done or completed. Slow, careful curation of documents occurs regularly. Care is taken to faithfully produce documents for future consumption, though not always for future implementation.

Caring for the documents can stand in for caring for people simply because the means to produce the paper and digital documents is available when resources to care for people are not. Sierra Leoneans working in government and private sector health organizations report that documents are the signs and symbols of preparedness that outsider governors care about most. In the health care organizations where our research team spent time, employees knew the language of documents and actively worked to provide the quotidian document ‘cover‘ for work activities. Tactics vary; several Freetown NGO leaders noted that the circulation of documents prepared by their colleagues often seemed dependent on the ‘find and replace‘ word processing function. One told of receiving a document that had many of the current health development buzzwords and terms – ‘innovative’, ‘gender’, ‘equity-based’, ‘influencers’, ‘sensitization’ – but the words were arranged nonsensically, in sentences that were impossible to follow. All the buzzwords were there, offered up as a ‘metacode’ (Rottenburg, 2009, 2014), a form of calculative equivalence that would qualify the NGO for resources from donors who cared about such terms. In such prose, health development norms are adhered and manifest. But such documents are ultimately empty vessels unable to meet universal healthcare needs, that is, providing healthcare for most of the people most of the time.

Since the end of the Ebola outbreak, there have been significant health systems improvements in Sierra Leone. Health security is widely talked about and understood as ‘work-in-progress’. Sierra Leoneans – from government officials to taxi drivers – are quick to report that actual global health security lies in local communities possessing an understanding of disease threats and formulating a competent first response. But alibi-making processes of epidemic preparedness documents continues and exists alongside Sierra Leonean ad hoc healthcare practices and systems improvements. Both Sierra Leoneans and expatriate global health actors are implicated. In my research, though, expatriate governors emerged as more often culpable for creating environments where everyday systems to care for the documents are usually more demanding than those caring for people. Documents, in all their banality, are usually overlooked as artifacts of significant health praxis. But they deserve a second look to determine if the work they do actually leads to lives saved, or just provide an implementation alibi.

Global health innovation: beyond hype

In November 2014, the WHO sent out a feature story: ‘Senegal’s recent stamping out of Ebola was achieved not only through its rapid infectious disease control work, but also by using a novel SMS-driven platform’ (WHO 2014). The WHO’s Short Message Service (SMS), sent over four million times, instructed people to

• Wash your hands with soap and water regularly.

• Avoid all contact with people who are sick with or have died from Ebola.

• Do not touch or eat the meat of dead or sick animals (monkey, rat, warthog, porcupine, pork).

Nowhere in the public WHO SMS materials is there acknowledgement that only 4 of 10 Senegalese women and 6 of 10 Senegalese men can read. This leaves about half of the adult population unable to read the messages that appear on their phones. Yet SMS health promotion is offered as a technical panacea to a major health problem, with no acknowledgment of illiteracy rates. Nor is there an acknowledgment that Senegal was able to fight Ebola exposures because of the long-term health system investments it has made over the last several decades, not because of SMS technology (see Murkovic, Thwing, & Diack, 2014).

‘Digital humanitarians’ advocate technology and big data analytics to support humanitarian relief, and there have been some successes (PLoS 2012). In-the-moment crowd-sourced digital geomapping, SMS language translation, and online emergency dispatch can work well. But global health technologies are not uniform and universal in their utility. In what follows, I elaborate on two innovations that did not work in Sierra Leone as advertised: 1) HealthMap’s claim of Ebola detection and 2) Harvard-based computational epidemiologists’ claim of Ebola containment via big data.

In the first example, Fast Company headlined ‘How This Algorithm Detected the Ebola Outbreak before Humans Could’ (Titlow, 2014). The originator of the algorithm was HealthMap (2014), an online global disease monitoring platform. It allegedly reported Ebola nine days before the WHO announced its presence in West Africa on 23 March 2014. The Associated Press went so far as crediting HealthMap as the ‘online tool [that] nailed Ebola’ (Associated Press, 2014). HealthMap co-founder John Brownstein is widely credited with the early detection and surveillance of Ebola (i.e. Wikipedia, 2018). But HealthMap’s algorithm, in fact, missed some of the very first notifications about ‘patients with Ebola-like symptoms’ because they were published in French and the algorithm was coded for English. Additionally, ‘[b]y the time HealthMap monitored its very first report, the Guinean government had actually already announced the outbreak and notified the WHO’ (Leetaru, 2014), making Health Map’s work as a detection tool, at the very least, redundant. But the biggest blow to claims of HealthMap’s algorithmic supremacy is the simple fact that a disease outbreak must first be reported – thus already detected – for the HealthMap algorithm to pick it up.

HealthMap is an interesting and valuable platform for visualizing global outbreaks of disease. But the heralded links between its algorithm, early detection, and improved health outcomes were not evident for the Ebola epidemic. HealthMap’s ability to simply note disease incidence did not directly translate to an ability to treat people with the virus, which was a complex challenge. Nor did it predict that Ebola would spread from its originating Guinean site to pandemic proportions across West Africa. HealthMap, after all, identifies disease outbreaks every day that do not become pandemics. HealthMap’s detection-outcome proficiency in the 2014–2016 Ebola outbreak was, in the final analysis, not profound. This does not take away from the value of a global disease visual like HealthMap, which is interesting in and of itself. It does, however, highlight a fetishization of a technology and an overstatement of its capabilities.

In the second example of unwarranted global health innovation hype, computational epidemiologists claimed that Ebola could be contained using big data analytics. They theorized that cell phone signals would leave a trail and thus create big digital data sets for tracking people who might be spreading Ebola. Cell phones would serve, according to this logic, as beacons of contagion, signalling where people infected with the disease are. There are several reasons why this did not work.

Firstly, it was the estimates of callers’ longitude and latitude coordinates that empowered the computational epidemiologists to think that by following signals from cell phone tower to cell phone tower, they would know where Ebola-sick people are. But these calculations were scaffolded onto a false first principle that cell phones are people. In Sierra Leone cell phones do not work well as beacons of individual identity and precise whereabouts because any given phone is not synonymous with an individual. Many Sierra Leoneans possess cell phones, but possession is often temporary. Cell phones are loaned, traded, and passed around among family and friends. In rural parts of the country, a single phone can be shared by an extended family, a neighborhood or a village. On the flip side of scarcity exists the numerous cell phones that many Sierra Leonean professionals have. Many people – from high-salaried professionals to high school students – have more than one cell phone. An HIV/AIDS counselor, Fatmata, had four cell phones, one for each primary phone network, because calling outside of a network was more expensive than having four separate cell phones. This kind of network access was also accomplished by having multiple SIM cards, rather than many cell phones. But Fatmata told me she liked separating the networks by phone to help her manage her various roles – as a health care worker, as a mother, as a friend, and as a small business owner. Another reason to have many cell phones, Fatmata pointed out, is how much time it takes to charge the battery. Electricity in her Freetown neighborhood is irregular, so one or two cell phones charge while she uses the others. With all these permutations, an actuality emerges: one phone does not equal an individual person, a point that upends a fundamental assumption of the computational epidemiologists who were interested in experimenting with the data from cell phones during an epidemic.

Secondly, the computational epidemiologists’ gravest mistake was presuming that the model of an endemic problem (malaria) would apply to an epidemic (Ebola). Cell phone data can only provide information about where millions of cell phones have gone, not if and where a person infected with malaria or Ebola is. Stopping the spread of Ebola depends on knowing the exact person in an exact location with the disease.

Further, the 2012 malaria model applied to Ebola used location and prevalence estimates. The model used 2008 and 2009 data long after the fact to estimate the locations of the nearly 15 million Kenyan cell phone subscribers. And, rather than actual cases of malaria, the model used 2009 estimates of malaria prevalence in targeted areas (Wesolowski et al., 2012, p. 268). To assume that the malaria model was sufficient for containment was to misunderstand the weaknesses of the correlated estimates themselves, as well as the kinds of containment activities that would halt Ebola’s spread. In Sierra Leone, it was contact tracing that brought the Ebola epidemic to an end. Thousands of Sierra Leoneans participated in contact tracing in Operation Os-to-Os (Operation House-to-House). Contact tracing is a labor-intensive, low-tech approach to containment, one that requires house-to-house visitations and face-to-face meetings with people who may be sick with Ebola or who have had direct contact with sick people.

Finally, any promise of Ebola containment by way of cell phones and big data is easily undone by another simple fact: Using cell phones to detect human mobility requires network coverage. Network coverage outside of Freetown is spotty, even though cell phone towers have sprung up across Sierra Leone over the last 15 years in remote rural locations. In 2014 in the area of Sierra Leone bordering Guinea and Liberia, as well as north and south of it – the site of the first outbreak and an area that experienced high Ebola mortality rates – thousands of people are regularly un- and underserved by cell phone coverage.

The argument that Paul Farmer has made about first-tier pharmaceuticals being made available to poor countries (Kidder, 2004) can also be made for global health innovation: Poor countries with complex health challenges should be able to access top-tier problem-solving technology. On-the-ground applicability can be a challenge, but pseudo global health as a critical analytic opens up a way to think beyond the incessant hype of global health innovation. Time and resource intensive experimentation during the Ebola crisis had opportunity costs which could have been avoided if, in the first instance, computational epidemiologists had not assumed malaria-Ebola commensurability and, instead, first asked West Africans – or anthropologists working in the region – about cell phone use.

Global health finance: dubious data

Development impact bonds, catastrophe bonds, and pandemic bonds are among a new class of financial instruments that aim to raise money for the everyday health, education, and human welfare needs of societies. These bonds are emblematic of a massive ideological shift in financing the world’s wickedest health problems: Human endeavors must now show ‘return on investment.’ In my research on health data and financing, I have found that unexpected things happen to health data in this performance-imperative milieu.

I have closely followed the colloquially named ‘Ebola bond,’ an integral component part of the World Bank’s Pandemic Emergency Facility (PEF). The PEF is a complex mechanism to finance pandemic response. Health data is used to ‘trigger’ the bond, that is, to prompt its payout for use in pandemic emergency response. But first, an explanation of the PEF itself: Beginning in late 2014 the World Bank brought over 60 diverse groups – governments, accounting and insurance corporations, lawyers and the like – together to create an instrument that would frontload money for quick future disbursement if a qualifying pandemic occurs. The aim was to shift pandemic emergency response away from donor country funding – which is considered unreliable (e.g. Grépin, 2015).

There are a lot of moving parts to the bond portion of the PEF, but the heart of it is this: In July 2017, 26 investors bought into a 3-year bond to raise US$320 million for pandemics. Some investors paid $US250,000, which was the minimum qualifying amount, and some invested $US50 million. If there is no qualifying pandemic through July of 2020, the high-risk tranche investors will get their money back plus about 12% interest per year. For the $US50 million investors, a 12 percent return is $US6 million each year, which is $US18 million over 3 years. If, over three years, a pandemic event does not happen, the $US50 million investors will get their $US50 million back plus $US18 million in interest. They will be paid this money if a pandemic event does not happen. The capitalist logic is that if investors are willing to risk their money on the chance that their money may be lost, then they deserve to make money if it is not called up in the service of a pandemic. According to this view, taking financial risk for a public health good should be well rewarded. But if there is a pandemic, health data would ‘trigger’ the money put up by the investors and it would be freed up for use in pandemic response. The data determines whether poor countries get the money or investors make the money. Making the point more poignant: The data does not need to be correct, it just has to work as a trigger to hold or release the investors’ money.

Payout for pandemic response is graduated. The first payout number is 250; 250 people need to have died or be confirmed sick with Ebola in two countries for the World Bank to release the first level of emergency response money, which is 30% of the investors’ money or $US 45 million. When 750 people die or are confirmed sick, 60% or $US 90 million is released. If the dead and sick reach 2500, then a final $US 150 million is released. The money will be split between pre-approved governments and ‘responding agencies’ (i.e., UN agencies and NGOs). More deaths mean more money for the response. There is a lot – of money, response capacity, and future number of lives saved – riding on the number of people counted as dead or sick. Some observers have insisted that the stakes are so high for investors that they will surely become more interested in supporting and financing health systems building and improvement.

In my research, though, investors and other governors have shown more interest in developing data collection, analysis, and visualization systems than in addressing the truculence of a health care system. Influencing data is possible from several sides. Higher death rate data mean that the bond money would be released for emergency response. Increasing death rates from a viral disease like Ebola is as easy as slowing down or stalling medical supply deliveries with simple customs concerns at transport facilities. On the investor side, deflating death rate data would be as easy as buying off the data collectors, restricting the gasoline supply for their motorcycles, or getting into their smart phones and tablets to alter data. When investors are gambling, when there is so much money to be gained or lost, there are a lot of creative ways to play with data triggers if that is what stands between gaining or losing a lot of money (see Kolhatkar, 2018). In the new forms of global health financing, data have become the lucrative deal makers and breakers in speculative finance, quite removed from the actual lived experiences of a well body.

Generating money for pandemic response is a common good. But as governments withdraw and drawdown aid, raising money for pandemic response has become more complicated. Will the Ebola bond work? Will investors consider health bonds good profit makers and continue putting money into them? And if they do, what will it mean to have private investors underwriting responses to public health risks, risks that have previously been backstopped by governments with constituencies?

Parting thoughts on faking and making global health

Impressive health gains were initiated in the mid-19^th century when governments orchestrated the organization of health care. Now austerity and privatization challenge health services worldwide. As a result, the world is at risk in largely incalculable ways as global health security, innovation, and finance practices move the world further away from embracing and encouraging government systems able to ensure health and well-being for most of the people most of the time. Governments are imperfect, but most are accountable to citizen constituencies (not customers) who might provide a check on ulterior motives of gross power, experimentation, or profit. This is not something private health provisioners are constitutionally obliged to do. In the high-pressure performance-imperative milieus of contemporary global health, cutting-and-pasting health preparedness documents, or heralding big data fixes, or championing financial instruments that gloss over the weaknesses of their data triggers, intimating global health improvements – but not actualizing them – makes a twisted kind of logical sense.

In global public health, magnanimous deeds and history-altering developments are commonplace. There have been many successes. But righteous works can catalyze ignoble means to ends, and the health sector seems always at a tipping point. Noble causes are vulnerable to corruptions (see Carreyrou, 2018) that can take promising initiatives off course. Appearances can deceive. Document alibis can hide what has not yet been accomplished in global health security; big data innovations can divert attention from essential emergency services during pandemics; and financial instruments can normalize the profitization of suffering – these are examples from three global health domains that deserve reckoning. On our current trajectory, health for all, or even most, is far from guaranteed. Eyes-wide-open critiques are fundamental to the continuing reduction of human suffering.

Acknowledgement

The author thanks the research participants; the Pseudo Global Health Working Group, especially Patricia Kingori and René Gerrets; as well as Earum Chaudhary, Musu Abdulai, Gary Parker, and Sam Eglin.

Disclosure statement

No potential conflict of interest was reported by the author.

Additional information

Funding

Government of Canada, Social Sciences and Humanities Research Council of Canada, Insight Development Grant #430 2012 0128,Insight Grant #435 2016 1076.