53V o l u m e 8 3 / 2 0 2 0 GLOBAL Introduction The US electoral cycle is one of the most closely watched political events in the twenty-first century. Indeed, in each successive year, new records are broken for expenditures on advertising by the campaigns. The digital sphere has become the main arena in which the various campaigns reach out to potential voters. By one 2019 estimate, “spending for political ads will reach $10 billion, an increase of 59% f rom the 2016 election year when an estimated $6.3 billion was spent. This represents a potential 16.5% of total local broadcast TV advertising revenue for 2020. Digital media is forecast for 21% of political ads, cable TV 14% and radio nearly 5%” (Adgate, 2019). This disproportional share for digital spending is indicative of what scholars have termed as the rise of computational politics, defined by one study as “the application of digital targeted-marketing technologies to election campaigns” (Chester and Montgomery, 2017: 1). Modern political campaigns, in the US and elsewhere, rely on the toolbox and data offered by technology corporations such as Facebook and Google, which are leaders in the online advertising industry. As explained by Wakabayashi GOOGLinG the US eLectOrAL cycLe, 2004-2016 South African Insights By Bhaso Ndzendze T H E T H I N K E R54 GLOBAL and Goldmacher: “The campaigns have been able to cater different messages for potential voters based on signals such as political leanings, what articles they have read, what videos they have watched and what things they have searched for. Instead of blanketing an entire city with a costly TV spot, the so-called microtargeting of political ads has become controversial because it allows advertisers to seek out specific voters and perhaps avoid broader scrutiny of their messages” (2019). In the wake of this, in October 2019, Twitter made the announcement that it would ban all political advertising from its platform. In turn, Twitter’s decision was widely interpreted as a response to the controversy caused by Facebook when it refused to de-platform a reportedly false video issued by the Trump campaign accusing a family member of a Democratic candidate of improper business conduct in Ukraine (Wakabayashi and Goldmacher, 2019). Twitter’s decision also led to Google’s own decision to do the same in November 2019. Many of these studies thus look at the US elections f rom within the US domestic context. Among internationally-minded studies, the interest is on the role of international dynamics on the actions of incumbents during the electoral cycle. For example, Chiozza (2015: 3) “assesses whether US presidents’ major responses in international crises reflect the variability in audience costs in an analysis of 66 international crises between 1937 and 2006”. The conclusion is that tying-hand commitment strategies (i.e., the creation of audience costs (potentially unpopular decisions) that will suffer after the fact if they fail to commit to the threat or commitment made (Fearon, 1997: 68)) were most preferred when presidential elections were approaching. This indicates that foreign policy is designed with re-electability in mind, lest unpopular foreign policy decisions lead to negative repercussions and loss of political office (Chiozza 2015: 3). The same findings on the foreign policy-electoral cycle nexus were identified by Gadarian (2010). Gadarian’s paper makes use of data over the 1980-2004 period to demonstrate that opposing political candidates in the US system are faced with different incentives for mentioning foreign policy during their campaigns. The paper effectively illustrates that American voters connect their own views on foreign policy when evaluating Republican candidates, while a lack or presence of similarity in foreign policy views has no impact in their evaluations of Democratic party candidates. Additionally, Gadarian’s paper proves that during times of external threat, US voters tend to not only be inclined towards hawkish candidates/ incumbents, but are even more likely to severely punish candidates who are perceived as holding dovish positions (Gadarian, 2010: 1046). In this regard, in 2016, CNN conducted interviews with 10 journalists f rom outside the US “for their take on the race so far, and what their country might be hoping for in America’s next president”. Analysing the responses, I note them to be mostly split between pro-Democrat (Canada, South Af rica, Iran and Japan) and neutral (United Kingdom, Venezuela, Israel, and Lebanon), with only two international journalists expressing indifference (India) and pro-Republican sentiment (Russia) (CNN, 2016: January 29). Among the participants was South Af rican journalist and noted editor of some leading national newspapers, Ferial Haffajee, whose contribution centred on the probability of a Trump victory: “Donald Trump? After Barack Obama? For those South Af ricans paying attention at this point in the U.S. presidential race, the primary campaign has prompted furrowed eyebrows. Indeed, the word “incredulous” best describes the response here to Trump’s howl-a-minute, holler-a-minute, horror-a-minute bid to become the Republican nominee” (Haffajee in CNN, 2016). She further expands, noting that: “Now the same country that elected Obama seems to be toying with the idea of electing a comb-over king who doesn’t seem to like Muslims and Mexicans very much, leaving some here to wonder what he feels about black Americans and Af ricans” (Haffajee in CNN, 2016). If subsequent Pew numbers are any indication, this may indeed be the case, as post- election approvals in South Af rica declined with In the wake of this, in October 2019, Twitter made the announcement that it would ban all political advertising from its platform. In turn, Twitter’s decision was widely interpreted as a response to the controversy caused by Facebook when it refused to de-platform a reportedly false video issued by the Trump campaign accusing a family member of a Democratic candidate of improper business conduct in Ukraine. 55V o l u m e 8 3 / 2 0 2 0 GLOBAL the election of Donald Trump. The second section is a breakdown of the methodology utilised in the article, as well as the electoral cycles of the US. The third section concludes the article, highlighting some areas for further research based on the findings. Methods This article traces Google queries in South Af rica over the 2004-2016 period for American presidential f rontrunners, eventual winners and incumbents. Within-country differences are also noted in this regard; the data will be broken into geographical concentrations, with the magnitude of each American candidate’s Google searches ranked by South Af rican provinces. Simultaneous worldwide Google searches will be used as a control group for each search trend per candidate to note whether interest in each candidate was on par, below, or above the global average. Importantly, this is a generative study and as such no proposal of cause-and-effect relations is made in this article. Rather, findings will generate working hypotheses and areas for further study. Election cycles in the US American presidential elections occur every four years and are dominated by two parties: the Democratic and Republican parties. With the system dominated by the two parties, their nomination contests can begin a year prior to the national election. Two of the party conventions of the election cycles took place in July of each year (i.e., 2004 and 2016), whereas in 2008, the parties had conventions in August (Democratic National Convention [DNC]) and September (Republican National Convention [RNC]). In 2012, the RNC was held in August and the DNC was held in September. Upon the parties nominating their candidates (along with their vice-presidential running mates), the party-to-party contest runs f rom their party conventions until the first Tuesday of November, which is the election day. As such, the dataset will capture and compare data for the electoral years f rom January to December of each year (i.e. 10 months prior to the election and one month after). Findings All data utilised in this article is obtained f rom Google Trends, a publicly available dataset of worldwide Google searches since 1 January 2004. All charts were generated by the author f rom sorted data. The Google Trends data are values that are calculated on an index that places scores f rom 0 to 100, “where 100 is the location with the most popularity as a f raction of total searches in that location, a value of 50 indicates a location which is half as popular” (Google, 2020). Importantly, a score of ‘0’ indicates a location where there was not enough data for this term. The timeline for both variables is 2004 to 2019. South Af rican searches for US presidential candidates In 2004, the South Af rican population searched for the Republican incumbent and eventual winner George W. Bush throughout the year. Whilst the Democratic nominee John Kerry surpassed Bush in March of 2004 when they declined, Google searches for Bush in South Af rica subsequently grew between April and July. In the decline in August, searches for Bush still outweighed those of Kerry. Google searches for Bush subsequently grew to new peaks in September, October and November respectively, before subsequently declining in December of that year. Noticeably, searches for both candidates reached their peak in November. South Af rica’s search f requency for George W. Bush during November was above the worldwide average, although the worldwide average search for John Kerry was above South Af rica’s in the same month. The majority of the searches occurred most in the North West province and the least in the Gauteng province. Bush’s share of searches was highest in North T H E T H I N K E R56 West (100%), and lowest in KwaZulu-Natal (KZN) province (at 60%, to Kerry’s 40%). In 2008, South Af rican Google searches were overwhelmingly for Barack Obama, the eventual winner of the election, though below the worldwide average until November, the month of the election which was once again the global peak. The majority of the searches for the candidates occurred most in the Gauteng province and least in the Northern Cape (NC) province. Barack Obama’s share of searches was highest in Free State (100%), and lowest in the Eastern Cape (EC) province (at 97%, to McCain’s 3%). In 2012, South Af rican Google searches were once again tilted towards Barack Obama, who was running for re-election, though the search f requency was now lower than the worldwide average. This diminished search f requency was also true of November, with the worldwide average outperforming South Af rica’s by a difference of 10 points. Worldwide searches for Mitt Romney were consistently higher than South Af rican searches for Mitt Romney, with searches for Romney outranking South Af rican searches for Obama in October 2012. This marks the only incident in the dataset in which a search for an unsuccessful candidate outranked the South Af rican searches for Barack Obama. The majority of the searches for the candidates in 2012 occurred most in the Gauteng province and least in the Northern Cape province. Barack Obama’s share of searches was highest in the NC province (100%), and lowest in the Western Cape (WC) province (at 96%, to Romney’s 4%). In 2016, South Af rican online searches for Donald Trump outranked those of Hillary Clinton for the entire course of the year. South Af rican searches for either candidate were below the worldwide average. Searches for both candidates peaked in November, though searches for Donald Trump matched the worldwide average, whilst searches for Hillary Clinton were below the worldwide average. The majority of the searches for the candidates in 2016 occurred most in the WC province and least in the FS province. Donald Trump’s share of searches was highest in the WC province (100%), and lowest in the FS province (at 91%, to Clinton’s 9%). Conclusion The paper has generated some insights in terms of the search trends of American presidential f rontrunners in South Af rica, using the worldwide search average as a control group. Most notably, all individual candidates who were the most GLOBAL 57V o l u m e 8 3 / 2 0 2 0 searched were the ones who went on to win the election. Nonetheless, there persist some areas for further research. Given variations in searches across the two countries, questions arise as to the causes of these. Do searches by either side act as proxies for forms of asymmetry, for example? More precisely, can we read into them notions of cultural hegemony, digital inequality, soft power, or diasporas on either side? Perhaps a combination of these may be at play. Perhaps none. Importantly, this need not coincide with traditional understandings of western preponderance as in these online dimensions, the global South can, as in no other sphere, exhibit patterns of some sovereignty. Yet, some scholars can read into the very ubiquity of Google as the preferred engine of the majority of the world as a sign of US dominance. Nonetheless, the modulation of search results to local geographies indicates that the relationship between Google and the consumer base outside the US is not unidirectional and its terms of operation are not dictated by the multinational corporation (Nahai, 2012). At the same time, within-country differences in search patterns also showcases the level of splintering of interest that should challenge any notion of homogeneity in South Af rican political perceptions of the United States. ■ References: Adgate, B. (2019). The 2020 Elections Will Set (Another) Ad Spending Record. Forbes [online]. Available at: https://www.forbes.com/sites/ bradadgate/2019/09/03/the-2020-elections-will-set-another-ad-spending- record/#287e1a9e1836 [Accessed 18 Jan. 2020] Aldrich, J.H., Sullivan, J.L. and Borgida, E. (1989). Foreign Affairs and Issue Voting: Do Presidential Candidates “Waltz Before A Blind Audience”? The American Political Science Review, 83(1), 123-141. Chester, J. and Montgomery, K. C. (2017). The role of digital marketing in political campaigns. Internet Policy Review, 6(4), 1-20. DOI: 10.14763/2017.4.773. Chiozza, G. (2015). Presidents on the cycle: Elections, audience costs, and coercive diplomacy. Conflict Management and Peace Science, 34(1), 3-26. CNN, (2016). From nervous laughter to Trump-supporting Kremlin: How world sees U.S. vote. CNN [online]. Available at: https://edition.cnn. com/2016/01/28/opinions/presidential-election-international-views- roundup/index.html [Accessed 18 Jan. 2020] Fearon, J. D. (1997). Signaling Foreign Policy Interests: Tying Hands versus Sinking Costs. The Journal of Conflict Resolution (Special issue: New Games: Modeling Domestic-International Linkages), 41(1), 68-90. Fishman, M. (2016). Russia: Why the Kremlin wants Trump. CNN [online]. Available at: https://edition.cnn.com/2016/01/28/opinions/presidential- election-international-views-roundup/index.html [Accessed 18 Jan. 2020] Fowler, E.F., Michael M. Franz, Gregory J. Martin, Zachary Peskowitz and Travis N. Ridoutk. (2018). Political Advertising Online and Offline [Working Paper]. Stanford University [online]. Available at: https://web.stanford. edu/~gjmartin/papers/Ads_Online_and_Offline_Working.pdf [Accessed 18 Jan. 2020] Friedman, U. (2017). 5 Ways to Interfere in American Elections—Without Breaking the Law. The Atlantic [online]. Available at: https://www. theatlantic.com/international/archive/2017/07/legal-ways-interfere- election/534057/ [Accessed 18 Jan. 2020] Gadarian, S.K. (2010). Foreign Policy at the Ballot Box: How Citizens Use Foreign Policy to Judge and Choose Candidates. The Journal of Politics, 72(4), 1046-1062. Gartner, C. (2016). The Science and Politics of Inf rastructure Research: Asserting Power, Place, and Agency in Inf rastructure Knowledge. Journal of Human Development and Capabilities, 17(3), 377-396. Haffajee, F. (2016). South Af rica: It’s Trump this and Trump that. CNN [online]. Available at: https://edition.cnn.com/2016/01/28/opinions/ presidential-election-international-views-roundup/index.html [Accessed 18 Jan. 2020] Jordan, M. (2015). How ratings-driven presidential debates are weakening American democracy. The Conversation [online]. Available at: https:// theconversation.com/how-ratings-driven-presidential-debates-are- weakening-american-democracy-50228 [Accessed 19 Jan. 2020] Kluver, R., Robert Hinck and Skye Cooley. (2016). How media outlets f rom around the world are reacting to the presidential campaign. The Conversation [online]. Available at: https://theconversation.com/how- media-outlets-f rom-around-the-world-are-reacting-to-the-presidential- campaign-66263 [Accessed 18 Jan. 2020] Nahai, N. (2012) English is dead, long live ‘glocalisation’. BBC News [online]. Available at: https://www.bbc.com/news/business-19801445 [Accessed 19 Jan. 2020] Perkins, D. (1957) Foreign Policy in Presidential Campaigns. Foreign Affairs [online]. Available at: https://www.foreignaffairs.com/articles/united- states/1957-01-01/foreign-policy-presidential-campaigns [Accessed 18 Jan. 2020] Wakabayashi, D. and Goldmacher, S. (2019) Google Policy Change Upends Online Plans for 2020 Campaigns. New York Times [online]. Available at: https://www.nytimes.com/2019/11/20/technology/google-political-ads- targeting.html [Accessed 18 Jan. 2020] Data sources US presidential candidate searches, 2004 (South Af rica). 2020. Available at: https://trends.google.com/trends/explore?date=all&geo=ZA&q=%2Fm%2F0c qt90,%2Fm%2F02mjmr,%2Fm%2F09b6zr [Accessed 18 Jan. 2020] US presidential candidate searches, 2004 (worldwide). 2020. Available at: https://trends.google.com/trends/explore?date=all&q=%2Fm%2F0cqt90,%2F m%2F02mjmr,%2Fm%2F09b6zr [Accessed 18 Jan. 2020] US presidential candidate searches, 2009 (South Af rica). 2020. Available at: https://trends.google.com/trends/explore?date=all&geo=ZA&q=%2Fm%2F02 mjmr,%2Fm%2F0bymv [Accessed 18 Jan. 2020] US presidential candidate searches, 2009 (worldwide). 2020. Available at: https://trends.google.com/trends/explore?date=all&q=%2Fm%2F02mjmr,%2 Fm%2F0bymv [Accessed 18 Jan. 2020] US presidential candidate searches, 2012 (South Af rica). 2020. Available at: https://trends.google.com/trends/explore?date=all&geo=ZA&q=%2Fm%2F02 mjmr,%2Fm%2F0271_s [Accessed 18 Jan. 2020] US presidential candidate searches, 2012 (worldwide). 2020. Available at: https://trends.google.com/trends/explore?date=all&q=%2Fm%2F02mjmr,%2 Fm%2F0271_s [Accessed 18 Jan. 2020] US presidential candidate searches, 2016 (South Af rica). 2020. Available at: https://trends.google.com/trends/explore?date=all&geo=ZA&q=%2Fm%2F0c qt90,%2Fm%2F0d06m5 [Accessed 18 Jan. 2020] US presidential candidate searches, 2016 (worldwide). 2020. Available at: https://trends.google.com/trends/explore?date=all&q=%2Fm%2F0cqt90,%2F m%2F0d06m5 [Accessed 18 Jan. 2020] South Af rican f rontrunner party leader searches, 2004 (worldwide). 2020. Available at: https://trends.google.com/trends/explore?date=all&q=%2Fm%2 F07q13,%2Fm%2F02tz36 [Accessed 18 Jan. 2020] South Af rican f rontrunner party leader searches, 2004 (US). 2020. Available at: https://trends.google.com/trends/explore?date=all&geo=US&q=%2Fm%2 F07q13,%2Fm%2F02tz36 [Accessed 18 Jan. 2020] South Af rican f rontrunner party leader searches, 2009 (worldwide). 2020. Available at: https://trends.google.com/trends/explore?date=all&q=%2Fm%2 F0466k4,%2Fm%2F0bzbzk [Accessed 18 Jan. 2020] South Af rican f rontrunner party leader searches, 2009 (US). 2020. Available at: https://trends.google.com/trends/explore?date=all&geo=US&q=%2Fm%2 F0466k4,%2Fm%2F0bzbzk [Accessed 18 Jan. 2020] South Af rican f rontrunner party leader searches, 2014 (worldwide). 2020. Available at: https://trends.google.com/trends/explore?date=all&q=%2Fm%2 F0466k4,%2Fm%2F0bzbzk [Accessed 18 Jan. 2020] South Af rican f rontrunner party leader searches, 2014 (US). 2020. Available at: https://trends.google.com/trends/explore?date=all&geo=US&q=%2Fm%2 F0466k4,%2Fm%2F0bzbzk [Accessed 18 Jan. 2020] South Af rican f rontrunner party leader searches, 2019 (worldwide). 2020. https://trends.google.com/trends/explore?date=all&q=%2Fm%2F04509v,%2F m%2F0n_8q8r [Accessed 18 Jan. 2020] South Af rican f rontrunner party leader searches, 2019 (US). 2020. Available at: https://trends.google.com/trends/explore?date=all&geo=US&q=%2Fm%2 F04509v,%2Fm%2F0n_8q8r [Accessed 18 Jan. 2020] GLOBAL