Border Economies/Capitalist Imaginaries: Dispelling Capitalism’s System Effects

Consider for a moment the three images below; each simple snap shots of international bridges between the U.S and Mexico that I took during my research. Look closely at the design features. Ever since the signing of the Treaty of Guadalupe Hidalgo of 1848, the U.S.-Mexico border has been located in the middle of the Rio Grande River. The borderline is marked on each of the international bridges via plaques and design features. These images, and the structures they capture, materialize the multiple mobilities that I want to consider. The bridges are at once made for crossing and yet, with only the slightest perceptual shift, they also manifest sovereign claims to national difference. The doubling of connection and difference is manifest in the mix of architectural unity co-mingled with design differences. The multi-vocality of borders, as it were, materialized in the form.

Figure 1

Figure 2

Figure 3

At times, demarcations are signaled through shifts in color: red to white in El Paso–Ciudad Juarez and subtle shifts from one shade of yellow to another in Brownsville–Matamoros. At other moments it is shape that signifies politically. On the Brownsville-Matamoros international bridge, the curve of the roof, the ridges, slats and struts all echo one another even as they differ. Perhaps most strikingly, the gap allows the two national bridge segments to connect, but not touch. Even here, when the bridges are held apart, the wire mesh behind the gap holds them together. The intricacies of bordering are materialized aesthetically allowing bridges to both connect and separate at the same time.

The doubleness of international bridges is not new to those who operate them. Luis Bazán, the Director of the Pharr-Reynosa International Bridge located one hundred and seventy-five miles east of Laredo, clearly understands bridging as a transnational undertaking. During his presentation to my colleague Laura Y. Liu and myself, Bazán stressed the importance of bridge expansion if Pharr was to maintain its edge in the intensely competitive market for bridge revenue (Bazán, 2017; Olaguibel, 2017; WorldCity (sic), 2020). Volume and wait times were front and center both of which, Bazán claimed, would be improved greatly by the building of new on-ramps, lanes, inspection stations, and warehouse facilities. But Bazán is no one dimensional nationalist; he works and thinks transnationally. Towards the end of the meeting, Bazán stressed the importance of collaborating with his Mexican counterparts as there was no point undertaking all of these ambitious developments, Bazán made clear, unless the Mexican half of the bridge developed at the same time. I began to see, that the border, for Bazán, was a condition of possibility, not a limit (Bazán, 2017). Similarly, border security companies take transnational operations as a given; politicians are the ones repeatedly drawing territorial lines (Border Security Expo, 2018; Aizeki et al, 2021).

What would it mean to have a politics located there, in the middle of the international bridges? How might we think capitalism from here? Before turning to duck-rabbit politics, two additional aspects of cross-border production must be introduced: bridges are only one element in the elaborate infrastructures needed to make cross-border production possible. Bridges are enmeshed in a larger set of production protocols, especially the in-bonding policies that generally fly under the political radar. While photographs capture the transnational aspects of bordering, they presume a level roadway crossing from one side of the river to the other–a condition that cannot be extended to cross-border production more generally. In fact, most sluicing mechanisms work to enable movement across uneven ground. Both aspects—the hidden and the uneven—need to be considered before I turn to politics in the final section of the article.

Global production and the long tentacles of networked production it entails, only functions if special structures and policies provide safe passage for the millions of parts to be assembled as they move through the ever higher, longer, more heavily securitized border walls. Supply chains and border walls can coexist because most international bridges identify special lanes, inspection stations, parking lots, and warehouses as in-bond spaces within which goods can cross the border duty free or at significantly reduced customs rates. Think duty free shops or old-fashioned transit lounges. Alternatively, in-bonding might be thought of as visas for things. Without safe passage for parts of products, or parts of parts, globalization would grind to a halt. In-bonding is a fine-tuned mechanism that allows international bridges to be simultaneously open and closed. These policies and the structures through which they are materialized are sluice gates extraordinaire.

In the latter half of the twentieth century, modes of production shifted with profound consequences for bordering. First, trade as a percentage of GDP increased enormously with trade as a percentage of GDP in the United States almost tripling from 10.75 in 1970 to 30.78 in 2011–with the percentage dropping to 26.31 in 2019 (World Bank, 2021). But the volume of trade itself was less consequential for my purposes than shifts in the mode of production. It is the ‘disintegration’ of production as Feenstra calls it (off shoring for others) that went along with increased trade that complicated bordering as older notions of the exchange of finished goods shifted to producing goods globally through multi-national supply chains (Feenstra, 1998; Cowie, 1999; Gereffi et al, 2005; Urry, 2014). The temporal sequencing of production and border crossing changed dramatically as Ricardian notions of comparative advantage in which goods were produced nationally and then traded—wine for cloth is the classic example—gives way. Under globalization the sequence of making and border crossing changes; now goods are produced across borders right from the start as supplier networks move parts to be assembled from around the globe. Over the last quarter century or more, as Pascal Lamay, the Director-General of the World Trade Organization from 2005–14, has said, goods are made in the world. Globalization has involved border crossing in fundamentally new ways (Lamay, 2014).

Take cars, for example. They are not made in Mexico and sold in the United States. Most products are made by moving people and things, or bits of things, back and forth across the border many times. Christopher Wilson from the Woodrow Wilson Center in Washington D.C. estimates that cars cross the border eight times while being made (Wilson, 2016). If anything, Wilson radically under-estimates the number of crossings in production now; there are approximately 30,000 parts to a car, and this is not taking into account the parts to make the parts. Thomas Klier’s and James Rubenstein’s amazing research on the U.S. automobile industry documents just how difficult it is to disarticulate supplier networks. The very idea of ‘domestic content’ as a distinct production category dissolves before one’s eyes. And this is to say nothing of the tools that go into building the parts. Parts, and parts of parts, are sourced from many countries through extensive supplier networks. Production occurs on top of or certainly back and forth across the border (Lamay, 2014; Wilson, 2016; Black et al, 2017; Patridge, 2016 & 2018). One quickly moves from the idea of supply chains, through networks, to what one of the people I interviewed described as ‘a multi-layered spiders’ web’ (Klier and Rubenstein, 2006; 2007; 2008; 2017; 2019a; 2019b; AA interview, 2018). It would be a mistake to assume that what counts as a part, or the country of origin, are themselves self evident classifications. Not so. Each of these aspects of global production are themselves deeply political.

Brexit has revealed the political nature of rules of origin and domestic content policies by revealing policies that had flown under the radar in pre-Brexit years. For example, see discussion of ‘aggregation rules’ by Campbell and Pooler (2017).

Crucially, not any old part of a larger product can travel in-bond, one has to apply for and be approved for safe passage. Classification is everything. One Brownsville engineer explained in-bonding as follows: compare windshield wipers and airbags. Both end up inside automobiles, but the path each takes across the border differs considerably. The wipers are considered end products and taxed accordingly, while airbags enjoy smoother passage as parts to be assembled and as such can cross the border via the in-bond lots (Edinbarough, 2017). The stakes are high; production costs vary considerably depending on whether items travel in-bond or not.

In-bonding is not new. It was introduced in the United States in the Smoot Hawley Tariff act of 1930. Sections 551–553 specified that goods passing through the United States to be sold elsewhere could be carried ‘by a bonded carrier without appraisement of payment of duties’ (Tariff Act, 1930). Passage of NAFTA in 1993 accelerated the dynamic by adding new incentives for the production of goods through extensive cross-border supplier networks (Wilson, 2016; Black et al, 2017). In-bonding reaches beyond ports proper to include the warehouses, customs brokers and expediting companies that serve as bonded agents throughout the United States. In Laredo alone, there are at least one hundred and thirty-eight customs brokers firms, though this likely is an undercount as brokers in other towns in the Rio Grande Valley also can serve as brokers in Laredo as well (Manta, 2021). Security at bonded warehouses is intense. The issue is not so much one of armed guards as it is extensive regulation and paperwork that make entering the business offices and bonded warehouses very difficult. Liu and I wanted to get a glimpse inside a bonded warehouse, to see how movement was materialized, but even the owner of the brokerage business would not let us in and was reprimanded sharply for even contemplating such action (Hattam, 2018). My point is not to deny that smuggling is part-and-parcel of cross-border trade; there is no doubt that it is of considerable significance (Keshavarz, 2018; 2019; Staudt, 1998). However, I want to keep my eye on formal mobility policies so as not to position mobility as outside of, or beyond, the state. This is not to deny informality, but to hold it at bay so as not to inadvertently homogenize state policy as focused on containment and the maintenance of border lines.

Who decides what items can travel in-bond? Through what processes? Based on what criteria? Answering these questions opens up critical aspects of bordering that have been obscured by the incessant media attention given to border walls. The Harmonized Tariff Schedule, established via the Omnibus Trade and Competitiveness Act in 1988, provides a comprehensive classificatory scheme for all goods entering the United States. As of February 2018, the Harmonized Tariff Schedule was 3,716 pages long and contained twenty-two sections and ninety-nine separate chapter headings. There are over seventeen thousand unique ten-digit classification codes in the schedule.

As with most policy innovations, the initiative had been in the works for years. President Ronald Reagan had requested that the US International Trade Commission draft the tariff schedule in 1981. After extensive review, the Omnibus Law was passed at the end of 1988 and went into effect on January 1st, 1989.

Disagreement is always a good place to start. Companies can contest particular classifications given to specific objects and somewhat surprisingly to me although U.S. trade policy is enacted through a myriad of national institutions, only one institution is responsible for deciding classification disputes: ‘Customs and Border Protection is the only agency that can provide legally binding advice or rulings on classification of imports.’

The Executive Office of the President identifies twenty-nine government trade agencies in the U.S. government. See Office of the United States Trade Representative, Executive Office of the President. 2017. ‘U.S. Government Trade Agencies’, available electronically at https://ustr.gov/about-us/trade-toolbox/us-government-trade-agencies. For the singular role of U.S. Customs and Border Protection and customs rulings, see the United States International Trade Commission, ‘Official Harmonized Tariff Schedule 2017’, available electronically at https://www.usitc.gov/tata/hts/index.htm. The language of exclusive jurisdiction has been softened in 2022.

The CROSS rulings are fascinating and immediately make apparent the ways in which customs classifications hinge on whether an object is classified as parts to be assembled or as an end-product. The boundary between the two is not always self evident. To grasp the dynamics of production across the border, I examined windshield wiper cases in the CROSS database from 1989 to the present. There were thirty-four rulings. I selected windshield wipers out of the myriad possible products because of the Trico wiper company has its North American manufacturing headquarters in Brownsville as well as a substantial manufacturing and assembly, packaging, and painting plant directly across the border in Matamoros.

Figure 4

The windshield wiper motor cases rest, in rather philosophical ways, on what exactly constitutes the essence of specific objects transiting across the U.S.-Mexico border. The case of HQ 964843 from January 30, 2002 puts the questions succinctly. The Daimler Chrysler Corporation had seven units of electric motors sitting in the Laredo, Texas Port of Entry that had been classified as end products under the 8501 heading of the Harmonized Tariff Schedule. Daimler Chrysler’s lawyers wrote to the Director of Customs National Commodity Specialist Division in New York on January 22, 2001 protesting classification of the motors as separate taxable objects rather than seeing them as parts to be assembled into the larger windshield wiper package under classification number 8512. Almost a full year later, John Durant, the Director of the Commercial Rulings Division, handed down the CROSS ruling on January 30, 2002 in which he denied the Daimler Chrysler petition and declared instead that the motors in question had been correctly classified under section 8501 of the Harmonized Tariff Schedule. The core of the three-page ruling declared that ‘the article designated in the 8512 heading text—a windshield wiper—is distinct from the motor that operates it. In our opinion, it is the wiper arm and blade, and not the motor, that imparts the essential character to the whole’ (CROSS, 2002). Echoes of philosophical debates over the axe head and handle reverberate. The stakes of the ruling were high; it determined whether the motors would be taxed or allowed to cross the border duty free. The decision was forwarded to Laredo for implementation. Windshield wipers were only one of many products whose identity was in question. Similar issues were raised for motors attached to seat belt assemblies, saw blades, car seat covers, and nut settlers. In each of these cases, the issue was whether the parts are permitted to travel in bond and tax free because they are parts to be assembled or whether they are end products and thus should be taxed at the port of entry (CROSS).

Windshield wipers are no exception: they too are produced across the border via global networks. However, the ways in which production straddles the U.S.-Mexican border generally flies under the political radar as extensive in-bonding infrastructures receive little attention, replaced as it were by continued reference to notions of domestic content. For example, both NAFTA and its recent USMCA revision claim that governments will measure and monitor domestic content down to specific percentage points. Proposals for identifying the point of origin of specific objects and their component parts requires projecting nationalist fantasies back onto global production networks. Highly integrated production processes cannot be disaggregated at the moment of border crossing. Look carefully at the windshield wiper motor shown in Figure 3. The object crosses the border as a unit whose seams are tightly sealed. The motor can not be easily opened for inspection and there is really no easy way to determine exactly where the parts that make up the motor were manufactured. Manifests serve as proxies for point of origin.

Perhaps the most compelling account of the difficulties of disaggregating nationally produced components can be found in Thomas Klier and James Rubenstein’s amazing research on the U.S. automobile industry. They document just how difficult it is to disarticulate supplier networks. The very idea of the ‘domestic content’ as a separate and readily identifiable category quickly dissolves before one’s eyes. And this is to say nothing of the tools that go into building the parts. One quickly moves from the idea of supply chains, through networks, to what one of the people I interviewed described as ‘a multi-layered spiders’ web’ (Klier and Rubenstein, 2006; 2007; 2008; 2017; 2019a; 2019b; AA interview, 2018).

Foregrounding cross-border production disrupts any easy invoking of the domestic. Separating here and there, near and far, foreign and domestic flies in the face of economic practices and has gone hand-in-hand with a cynical nationalist politics.

A striking aspect of the Rio Grande Valley is the ways in which cities and towns often are paired along the border: Brownsville-Matamoros; McAllen-Reynosa; Laredo-Neuvo-Laredo; El Paso-Ciudad Juarez; Calexico-Mexicali; San Diego-Tijuana among many others. Beginning in the 1980s, many multi-national corporations sought economic advantage by manufacturing across the U.S.-Mexico border through what was then known as the ‘twin plant’, or maquiladora production model. The operating premise of maquiladora production was that costs could be arbitraged across the border with assembly work being done in Mexico while management, distribution—and profit—would be located in the United States or in some other global capital. In fact, wages were only one differences to be arbitraged: production adjacencies, water rights, environmental regulations, health and safety, trade treaties, and labor laws all are considered.

Labor costs should not be over estimated when it comes to globalization as labor is a smaller percentage of the overall cost of most items than popular media accounts often suggest. The maquiladora literature is enormous; Classics include Patricia Fernandez-Kelly (1984), Vicki L. Ruiz and Susan Tiano (1987), Leslie Salinger (2003), Carolyn Tuttle (2012).

I went to the border with accounts of maquiladora production in my head only to find that my presumptions about where and how inequalities worked were insufficient. Circuits of power and inequalities were intense, but not in ways that I had anticipated. Two of my interviewees noted that differences between Texas and Mexican labor laws also were important factors in maquiladora production. It took me a moment to understand exactly what they were suggesting. I had presumed, as we are encouraged to do in the United States, that as an ‘advanced industrial society’, the United States would have the more robust labor protections. My interviewees pointed out that this is not always so. In fact, one of the advantages of the twin-plant model for multinationals was the ability to leverage Texas’s right-to-work laws during the global production. I had not seen this coming. I had not been attending to Texas labor law at all. Once my interlocutors drew my attention there, the dynamic relation between Mexican and Texas labor regulation in which corporations could play different aspects of labor vulnerability off against each other began to make sense. Seen in terms of labor rights, Texas workers received little or no protections under the right-to-work laws that have been in place since 1993. Mexican labor law, by contrast, includes several important provisions that at least on face value provide important protections to Mexican labor. Three provisions are particularly notable: termination, maternity leave, and profit sharing. Although wages are indeed much lower in Mexico than in the United States, termination is much more difficult once a thirty-day probationary period has passed. In addition, Mexican law stipulates twelve weeks maternity leave must be provided. Finally, the profit-sharing provision requires employers to distribute ten percent of their taxable profits to workers each year.

For succinct overviews of Mexican labor law, see Jorge Millan, Mario Perera and Jon Lowe (1990), and Jorge A. Silva Rodriguez de San Miguel (2020). For an example of consultant advice to U.S. firms on how to navigate Mexican labor law, see The Offshore Group, “Mexican Labor Law: An Introductory guide to Mexican Labor Laws and how to navigate them,” no date. In author’s possession. For discussion of contemporary efforts to strengthen Mexican labor law, see Christine Murray (2022).

Having laws on the books is one thing, implementing them is quite another. Many acknowledge that companies employ all sorts of tactics to avoid aspects of these laws–especially the requirement to share profits. When pressed, one interviewee explained that, on occasion, company executives would schedule meetings on the U.S. side of the border in part as a way of avoiding Mexican labor laws and to shift profit centers south to north. Global entry visas (and their precursors) allow managers make this strategy by minimizing wait times at the ports of entry (AA, 2018; and JC, 2020). Multinational corporations thus arbitrage labor laws much as they do other components of global production. The unexpected aspect was the fact that when it comes to labor law it was Texas, rather than Mexico, that had the weakest labor protections. Arbitrage thus works in both directions. I know nothing in the globalization literature that examines the dynamic relation between the right to work laws in Texas and the cost of labor in Mexico. A duck-rabbit perspective actively works to keep both sites of inequality and exploitation in focus. This was arbitrage of a different kind and in a different direction than I had anticipated: Texas, rather than Tamaulipas, serves as the Delaware of cross-border production; Texas is the state that has eviscerated labor protections. The globalization literature has a lot to say about labor exploitation and outsourcing, but I have seen no discussion of this cross-border braiding of labor exploitation described by my interlocutors.

For some, the Texas right-to-work laws might be seen as adding an extra loop in capitalism’s rapacious reach without fundamentally altering capitalism’s enduring exploitation of labor. In the last analysis, one might say, labor always is the source of profit. The specific route taken varies, but the underlying logic is the same. I want to suggest viewing cross-border labor law this way is a mistake; it provides too flat a reading of the labor politics in play. Yes, corporations exploit labor on both side of the border, but reading the central dynamic as exclusively, or even centrally, as the principle lens through which exploitation occurs obscures the ways in which right-to-work laws were entangled with the broader conservative political project that reached well beyond capitalism. As Joseph Lowndes, Marc Dixon and others have shown, conservatism in the United States is forged out of many elements, including the increasingly prominent right-wing extremists. The different elements did not have a natural political affinity one with the other; they were not just lying around as it were waiting for the dots to be connected. On the contrary, considerable political labor went into the forging of new identifications; the affinities had to be made and remade, yoked together into enduring political formations. Joseph Lowndes has tracked the unexpected assemblage of political identification on the right from the 1940s through the present while Marc Dixon considers the assemblage of disparate political elements supporting the right-to-work laws and William Connolly examines the evangelical–capitalist resonance machine (Lowndes, 2009; 2022; Hattam and Lowndes, 2013; Dixon, 2012; Connolly, 2005). The role of imagination and creativity in the making of affinities is not to be underestimated. The critical aspect of the duck-rabbit image is the way the figure draws our attention to the plasticity of perception. Creativity does not just bring the elements together, the reorganization allows elements to be seen that were not previously visible. Imagination and evidence go hand in hand enabling new political dynamics to emerge (Dunne et al, 2022).

The scale of industrial parks and multi-national factories serves as a powerful reminder that cross-border production is deeply embedded in extracting the benefits of low cost labor. But when seeing maquiladora production in Reynosa, the right-to-work laws were not in my field of vision. The continued focus on outsourcing as the central dynamic of globalization deflects attention from work and labor politics inside the United States (Urry, 2014). By repeatedly looking abroad, by attending to the ways in which multi-national corporations move production abroad to reap the benefits of cheap labor, leaves exploitation at home, in Texas, out of sight. Apple-Foxconn generally are used as the models of globalization, with design occurring in Cupertino and production in Shenzhen. On this telling, inequalities associated with globalization are located elsewhere–overseas. James Fallows widely read article in the Atlantic Monthly is the classic case in point (Fallows, 2007). The domestic continually is positioned as the site of privilege; Right-to-work laws in the United States are not mentioned. Examining cross-border production in the Rio Grande Valley (RGV) drew my attention to inequalities within, to Texas as a site of limited or non-existent labor protections. And in so doing, the RGV has made me aware of the ways in which the language of outsourcing that pervades discussions of globalization deflect attention away from inequalities within.

One last step in the argument is needed. Attending to misalignments and the sluice gates that bridge them allows us to glimpse the fantasies of coherence that pervade everyday as well as scholarly accounts of capitalism. Drawing on Roland Barthes and Timothy Mitchell, I am suggesting that sluice gates allow us to glimpse capitalism’s ‘system effects.’ For almost half a century, the term global supply chains has carried with it notions of linear connectivity that transforms capitalism’s heterogeneous elements into an apparently coherent production system. But as Covid-19 has made clear, supply chains are easily ensnared. Blockages accrue. To be sure, at times fissures are acknowledged, but in the last analysis the logics of capitalism are presumed to be determinative. Claims to coherence is in some sense what makes it a system; ultimately, the pieces have to work together or else the object of analysis morphs into an assemblage or other form of social or political formation.

Looking from the middle of the bridge—from the sluice gates perspective—gives the misalignments greater analytic force. One’s eye settles on the disjunctions and on the extensive imaginative and political labor needed to bridge the gaps. Underlying affinities among the parts are no longer presumed; rather the affinities have had to be made and remade both imaginatively and materially. The sluice gates give testimony to the precarity of the seams within. The so-called system is shot through with misalignments. System effects draw one’s eye away from the partial connections to dwell as it were on the completeness of the capitalist project. An imaginative infrastructure is needed for coherence to appear as capitalism’s robust and expansive aspect. Sluice gates recede as the system effects capture the imaginative space.

Put differently, capitalism’s specter pushes sluice gates aside leaving exploitation as the key analytic. By intentionally placing sluice gates at the center, by looking at capitalism from there, from the middle of the bridge, brings capitalism’s system effects into the field of vision. Once fantasies of capitalist coherence—and its anti-capitalist reflection—have been relinquished, new economic and political imaginaries emerge. While the specter hovers, imaginative spaces close in.

Capitalism is not a duck. Capitalism is not one thing. Multiple mobilities, misalignments, and looping circuits of power abound. I have tried to disrupt capitalism’s system effects by attending to the gaps within and by considering the ways gaps must be bridged imaginatively and materially for global supply chains to function. Focusing on sluice gates allows us to see capitalism in a duck-rabbit way as simultaneously pervasive and incomplete. Cross-border production networks depend on international bridges, in-bonding systems, and uneven labor laws to work. Each of these mechanisms allows global supplier networks to co-habit with ever-stronger claims to national sovereignty and border security. In fact, cross border entanglements are enacted on a daily basis, but generally are not included in the accounts we give of how globalization functions. Notions of domestic content and outsourcing that infuse debates over globalization allow nationalist fantasies to be projected back onto integrated production processes as if the foreign and domestic could be kept apart. I am interested in considering a politics from within the entanglements, from in amongst the supply chains and cross-border production networks.

If system effects can be relinquished, the horizon of political possibilities shifts. Exploitation and inequality persist, but the eye can be drawn to the innumerable ways in which capitalist logics remain incomplete. Critiques of capitalism abound, but almost all take their orientation from capitalism as we know it. An imaginative counter politics seeks a broader horizon that pivots off the gaps, misalignments, and reversals. Centering the imaginative aspects of politics is especially pressing now as old cartographies are faltering. Production practices exceed the analytics at hand. The anti-work movement is a perfect case in point: what are we to make of the fact that 1.8 million people have signed up for the reddit antiwork thread in which the by line states, ‘Antiwork: Unemployment for all, not just the rich’ (reddit, 2022). Some have suggested that the Great Resignation is really a misnomer, that the shifts are more churn than anything more significant (Thompson, 2021). This seems to be too narrow an analytic that likely misses the dispositional shifts at hand. The contemporary anti-work movement is at once churn and resistance, but most interestingly for me, is the question of whether it is also a harbinger of more wide-ranging changes in economic and political imaginaries.

Imaginative politics is not without its liabilities. The sure footedness that accompanies both an anti-capitalist and informal politics critiques has to be relinquished, since the issue of orientation is precisely what is being put into question. Long standing assumptions about the directionality of movement and what counts as higher and lower ground do not hold. The guideposts that have been used to orient our politics for over a century are unsettled. But this is not necessarily a negative; it is the unsettling that matters. Disorientation, if we can tolerate it, is where the imagination lies. If one always knows where one is headed, if the terms of engagement are known in advance, then inevitably the domain of the political is diminished. An imaginative, as opposed to a critical, politics involves accepting the risk of not knowing quite where, or with whom, one stands. Where a critical politics demands that we know which side we are and makes questions of identification and mobilization the central pillars of an anti-capitalist politics. The duck-rabbit politics insists on relinquishing that demand, on relinquishing the all-knowing voice that simultaneously authorizes and narrows the political possibilities at hand. The politics that I am after are not so neatly contained, the remits not so tightly pre-scripted. I want an imaginative politics that centers creativity rather than a critical anti-politics. The world is a wild and interesting place that carries within it enormous opportunities for the terms of political engagement to be imaginatively reconfigured.