Rising With the Robots: Towards a Human-Machine Autonomy for Digital Socialism
Christopher M. Cox
The University of Akron, Akron, OH, USA, ccox@uakron.edu
Abstract: This essay is concerned with conceptualising digital socialism in two ways. First, this essay typifies digital socialism as a real utopian project bringing together the utopian potential of “full automation” as tied to socio-economic imperatives indicative of socialist aims. Second, in recognition of a critical gap between full automation and an emerging technological autonomy, this essay argues for a human-machine autonomy that situates autonomy as a shared condition among humans and machines. By conceiving of humans and automated technologies as autonomous subject aligned against capital, pursuing the aims of digital socialism can anticipate and avoid capitalist ideologies that hinders possibilities for autonomous pursuit of digital socialism.
Keywords: digital socialism, robots, robotics, fully-automated communism, full automation, human-machine autonomy, Autonomist Marxism, Autonomism
This essay
attempts to answer the question “how do recent debates about ‘full automation’
and postcapitalist socioeconomics establish a
foundation for conceptualizing digital socialism?”
Since 2015, debates about socialism have exhibited a resurgence among
public consciousness and electoral politics in the western world. From the
election of “Marx-admiring socialist” (Danner 2015) Jeremy Corbyn as leader of
the UK Labour Party to successive Bernie Sanders U.S. presidential campaigns
foregrounding “democratic socialism” as a series of comprehensive
socio-economic reforms aimed at creating an economy that “works for all, not
just the very wealthy” (Frizell 2015) to the surge of 13,000 new members and 100 new chapters
of the Democratic Socialists of America (DSA) (Schwartz 2017), “socialism” is increasingly mobilized by political
actors.
Concurrently,
the emergence of Jacobin as the
“leading intellectual voice of the American left” (Matthews 2016) and the publication of The Socialist Manifesto: The Case for Radical Politics in an Era of
Extreme Inequality by Jacobin
founder Bhaskar Sunkara (2019) – and its coverage in mainstream news outlets from Wall Street Journal (Swaim 2019) to Slate (Weissman 2019) – have contributed to “more interest in – and support
for – socialist ideas than at any time in recent American history” (Nichols 2015, xxv).
Alongside
increasing socialist consciousness and mobilisation, the latter part of the
2010s gave rise to the era of “automated connectivity” (Van Dijck 2013, 23), as automated processes built into digital platforms
and techniques became significant forces in the production, governance, and
maintenance of social life (Gillespie 2014; Bucher 2018). Since 2015, in tandem with the increasing interest in
socialism, “full automation” has become increasingly central to imagining life
beyond capitalism and thinking through the material means of reconstituting the
production and provisioning of labour, goods, and services.
To better understand the relationship between
contemporary automation and socialism, this essay teases out critical strands
among recent scholarship to, first, define the imperatives of “digital
socialism” and, second, interject concerns for technological autonomy otherwise
neglected among full automation debates. In doing so, the essay argues for
digital socialism to be understood as a “real utopian” (Wright 2010) project, outlines the imperatives of digital socialism,
and stresses the opportunity to conceive of autonomy as a shared condition
among humans and machines to better solidify and anticipate class solidarity
amidst the struggle to achieve socialist ends. The essay proceeds as follows:
· Digital
Socialism: Full Automation as Real Utopia (section 2)
· Socialist
Imperative 1: Shifting Values and Ethics Associated with Labour (section 3)
· Socialist
Imperative 2: Centralised Planning (section 4)
· Socialist
Imperative 3: Basic Services (section 5)
· Digital
Socialism: Strategic Imperatives and Critical Opportunities (section 6)
· Dualities
of Autonomy: Oppositions Between Human and Technological Autonomy (section 7)
· Becoming
AutonomoUS: Human-Machine Autonomy (section 8)
·
Human-Machine Autonomy and Solidarity Against
Capital (section 9)
Socialism is a
socio-economic system predicated on maximising cooperation, democratic participation,
and egalitarian outcomes in all spheres of life. Communism is a successive
stage of the socialist project that enables all property and means of
production to be held in commons (i.e. communally-owned), basic necessities to
be apportioned based on a person’s needs, and a centralized source (often
conceived as the government) to maintain the mechanisms for communal ownership
and equitable distribution. In and outside the United States, the terms
“socialism and “communism” tend to be equated with Maoism, Stalinism, and other
national communisms that installed totalitarian ideologues who amassed and
wielded power through a form of “state capitalism” (Sperber 2019) ill-suited to ennobling the
proletariat and resolving class struggle through democratic participation. On
this front, the contemporary promulgation of “socialism” is at least somewhat
attributable to the fact that it is no longer subject to “the burden of being
associated in the propaganda systems of East and West with Soviet tyranny” (Chomsky 2016), a circumstance that enabled the East to
maintain power through the aura of socialist aims and the West to demonize
socialism and communism outright.
Socialism has led been associated with utopian thought, leading Engels to distinguish between utopian socialism (prominent in the 19th century) and the historical materialism of Marx’s “scientific socialism” (Engels 1880). Utopia is a broad term that encompasses a range of ideas about idealised conditions for society. In most cases, utopia is a place, one that is less a physical destination and more of a hypothetical realm where a harmonious society can be realised. Utopianism, in this sense, is a mode of thinking that attempts to imagine the conditions enabling social harmony, particularly in the context of governmental and economic relations. Utopian socialists in the 1880s attempted to foster social harmony by “devising plans to make society more cooperative, production more efficient, and distribution more fair” (Paden 2002, 68). These plans did not account for class politics and struggle and thus, for Marx and Engels, failed to represent the interests of any class, much less the proletariat forced into inevitable contact with capitalist ownership (Paden 2002, 68). For Engels (1880), then, “scientific socialism” entails similar ideas about cooperation and equity rooted in “the materialistic conception of history and the revelation of the secret of capitalistic production through surplus-value” (305) such that socialism is primarily concerned with “the necessary outcome of the struggle between two historically developed classes – the proletariat and the bourgeoisie” (304).
Others attempt to mitigate the ethereal qualities of utopianism with a more pragmatic perspective on concrete political relations. Notably, Ernst Bloch focused on the spiritual aspect of utopia and the ability for political change to alleviate material burdens that undermine spiritual fulfilment in lived experience. For Bloch, utopia entails “the world of the soul, the external, cosmic function of utopia, maintained against misery, death, the husk-realm of mere physical nature” (Bloch 1918/2000, 3) [emphasis in original]. Bloch’s utopianism is not an exotic hideaway or a proverbial Shangri-La but an inward journey of reflection and recognition of the universal conditions underlying the disillusion and disaffection experienced by many in modern capitalism. By reflecting on this spiritual condition of modern life, the journey inward illuminates the potential for a “utopian reality” (Bloch 1918/2000, 179) envisioned as the ability to strive for a realm of fulfilment capable of being actualized in the material world. The recent emergence of a “real utopia” works from a similar register.
As devised by Erik Olin Wright, a “real utopian” project seeks practical opportunities to restructure social institutions to instantiate alternatives to capitalism and materialise “radical democratic egalitarianism” (2010, 22). By anchoring the large-scale optimistic imaginings of utopian thinking with “specific proposals for the fundamental redesign of different arenas of social instructions” and “immediately attainable reforms of existing practices” (Wright 2010, ii), real utopianism pursues new models for egalitarian practice predicated on their viability and achievability. While Marx and Engels criticise utopian socialists for a reactionary tendency to appeal to working masses with a religious zeal that harkens towards “castles in the air” rather than grounded political struggle (1848, 516-517), the real utopian project squares the circle by situating utopian visions as an outward projection from the fruits of viable material intervention. In doing so, discourses around full automation conceive of automated technologies as “vectors for new utopias” (Hester 2018, 8) and oppose the seeming unfeasibility and immobility of revolutionary traction, calling for “the futural orientation of utopias” combined with “real tendencies of the world today” to devise a feasible starting point for life beyond capitalism oriented towards continued progress and development (Srnicek and Williams 2015, 108). In this way, contemporary ideas about full automation epitomise the spirit of real utopianism by, first, specifying contemporary conditions that make economic and social reconfigurations strategically viable and, second, leverage “full automation” as frame for utopian imaginings.
As a real
utopian project, the emerging digital socialism seeks to avoid the
technological determinisms often associated with the
“California ideology”) (Barbrook
and Cameron 1996; Turner 2010) and other
similar utopian ideas that stress free enterprise and marketplace expansion as
the primary means of creating innovative technologies poised to change society
for the better. Instead, real utopianism prioritises structural changes to
policies and ideologies about the interrelationships of work, ownership,
resource planning and allocation. By typifying the specifics of digital
socialism as it emerges from debates about full automation from 2015-2019, the
goal here is to clarify the lay of the land as it is, rather than argue for
what it should be. Key works emerging during this timeframe include Inventing the Future: Postcapitalism
and a World Without Work (Srnicek
and Williams 2015), Four Futures: Life After Capitalism (Frase 2016), The Automatic Society: The Future of Work (Stiegler 2018), Xenofeminism (Hester 2018), Fully-Automated
Luxury Communism (Bastani 2019),
and Inhuman Power: Artificial
Intelligence and the Future of Capitalism (Dyer-Witheford, Kjøsen and Steinhoff
2019). All of these monographs in their own
ways and to varying degrees – engage with full automation in conjunction with
describing socio-economic conditions necessary to imagine a postcapitalist
world reflective of socialist aims (whether or not the term “socialism” is
explicitly used). While I ultimately take up the task of more forcefully
arguing for a solidified conception of human and technological autonomy, the
first step is to collate and clarify the specifics of digital socialism as it
currently stands. In doing so, I identity and specify three socialist
imperatives significant to current ideas around full automation:
· Generating
New Ethics, Values, and Arrangements for Labour
· Centralising
Economic Planning
· Implementing
Basic Services
In the case of the first socialist imperative, Srnicek and Williams (2015,
125) decry the capitalist work ethic’s insistence that “renumeration
be tied to suffering” and suffering the indignities and inequities of
capitalist exploitation is “the only means for true self-fulfillment”.
To their minds, the ability to implement full automation must account for the
long-standing desire to attain status through work, even if said work is seen
as undesirable. Significant cultural shifts around the work ethic can not only
help underscore the possibilities for personal and collective fulfillment beyond market demands but also mitigate the
precarity and turbulence of labour markets increasingly apt to diminish the
abundance, variability, and remunerative sufficiency of work prospects. As a
viable starting point, shortening the hours of the formal work week could
reduce the amount of hours in the five-day week or institute a permanent
three-day weekend (Srnicek
and Williams 2015, 116).
In a similar vein, xenofeminism situates automation and other technologies as the means to reconfigure cultural notions of work. Xenofeminism is a recent strand of technofeminism that seeks to abolish the imposition of gender binaries and overturn essentialist gender ideologies associated with biological and social reproduction (Hester 2018). Xenofeminism’s pursuit of gender abolition and anti-naturalism stresses “post-industrial automation” (Hester 2018, 8) and related techno-materialisms as a means of changing concrete relations among gender, work, and social institutions (such as the family). Understood as a “multiply gendered world” (Hester 2018, 30), xenofeminsm’s gender abolition seeks to enlarge the range of gendered expression and concomitantly undo expectations of domestic labour as tied to gender. Shifting cultural ideas about the constructed and contested nature of gender is part and parcel of undoing “culturally weaponised markers of identity that harbor injustices” (Hester 2018, 30), including the necessity of child labour as a potentially dangerous bodily labour undertaken by women and ensuing expectations around childrearing and domestic caretaking. Beyond undoing gender naturalism and expectations around domestic and biological labour, Hester offers Donna Haraway’s (2016) notion of “kin” as a “means of prioritizing the generation of new kinds of support networks” (Hester 2018, 63) to realise new forms of collaborative work and care, a “counter-social reproduction” envisioned as “social reproduction against the reproduction of the social as it stands” (Hester 2018, 64) [emphasis in original].
Offering full automation as an orienting force for
collectivist formation and harmony, Bernard Stiegler
(2018) situates capitalist automation as an
entropic force that generates ever-increasing uncertainty, disorder, and
instability in capitalist markets and society at large. These sentiments are
internalised by everyday people who come to envision the future as inhospitable
and devoid of the potential for widespread prosperity. For Stiegler
(2018, 7), full automation is an opportunity for
“dis-automatization”, the harnessing of energies previously-dedicated to formal
wage labour and recalibrated towards “collective
investment of the productivity gains derived from automatization” (Stiegler 2018, 15) [emphasis in
original], a collectivism predicated on a negentropic
perspective that offsets the disarray of capitalism’s automated entropy and
engenders possibilities for egalitarian order and harmony. “Collective
investment” also speaks to the need for economic and personal investment in
work arrangements to be understood in terms of cooperation and plurality,
eschewing the individualistic drives of capitalism to amass capital in defiance
of the common good.
In their critique of full automation that foregrounds the role of artificial intelligence (AI), Dyer-Witheford, Kjøsen and Steinhoff’s (2019, 153-156) “communist orientation to AI” attempts to reconfigure the relationship between postcapitalist futures and automated technologies. Rather than positing full automation as an opportunistic moment or tool to be seized and applied towards the break from capitalism, a communist orientation to AI prioritises “liquidating the structural dynamics of capital” (153) undergirding the development of automated technologies and the ethos built into these technologies by companies driven to patternise forms of social interactivity antithetical to socialist solidarity. To these authors’ minds, the goal should be to expropriate capital from AI, collectivise ownership of AI, and retrain AI to function in accordance with collectivised values and structure to enact a “true democratization of AI” (Dyer-Witheford, Kjøsen and Steinhoff 2019, 154). Among such collectivised values, collectivised ownership is a central facet of the second strategic imperative of digital socialism.
Similar to the way “full automation” is situated as a
touchstone for charting a course beyond the confines of capitalism,
“post-scarcity” is a similar touchstone among advocates of centralised planning
(Frase 2016; Bastani 2019; Phillips and Rozworski
2019) who attempt to reconcile the technological ability to produce an
uncapped abundance of goods and services with the artificial limits placed on
production and dissemination by private ownership. Central planning entails a
production of goods and services as directed by a governmental source to
equitably allocate these goods and services. Even as some aspects of
centralised planning receive more analysis than others, one of the recurrent
themes among central planners is the ability to generate an abundance of goods
and services equitability doled out through such planning, thereby transcending
scarcity as a circumstance of private accumulation. One of the common threads
about central planning advocacy is that forms of privatised central planning
already in place are useful structures capable of being refashioned to
equitably allocate resources.
One of the
central tenets of Aaron Bastani’s (2019) advocation for a
fully-automated luxury communism is the necessity of demanding “the
intentional, conscious planning at the heart of modern capitalism be repurposed
to socially useful ends rather than socially destructive ones” (227). Bastani shows a
particular concern for finance, foregrounding centralised banking and
“municipal protectionism” (207) as two interrelated
facets of central planning. Whereas central and private banks currently
prioritise the administration of loans and other fiduciary mechanisms based on
the assets held by a borrower and the likelihood of lending as a profitable
venture, nationalised central banking shows the potential to guard against the inequities
of capitalist finance by overseeing an informal network of locally-owned
businesses and banks. In this context, “municipal protectionism” refers to the
pursuit of localised businesses owned by workers that can better mitigate
inequality through a wider range of ownership models (210).
Tied to finance emanating from local banks and credit unions and mitigated by a
centralised national bank tasked with ensuring equitable allocation, these
types of businesses can protect against micro forms of capitalist domination
within the workplace and macro forms of capitalist domination in society at
large (211).
In Leigh
Phillips and Michael Rozworski’s (2019) monograph-length case study of
Wal-Mart’s internal central planning, they stress “openness and cooperation
along the supply chain” as fundamental to planning and the ability to
continuously replenish resources (2019, 38).
By arguing for Wal-Mart’s internal supply chain structure as a centralised mode
of production and distribution predicated on collaboration and cooperation
among participants, Phillips and Rozworski highlight
institutional and political practices already at play that offer a viable means
of recalibrating towards socialist ends. In their conception, central planning
is not the purview of a small group of planners, programmers, or algorithmic
calculations but, instead, relies on democratic participation at all points of
production and consumption, if “computer-assisted, decentralized, democratic
economic decision making” is to be realised (Phillips
and Rozworski 2019, 213).
By treating
the possibility of full automation and (post)scarcity as a given, Peter Frase (2016) outlines four
possible future scenarios based on recomposed dualities of dualities of
hierarchy/egalitarianism and scarcity/abundance: communism (egalitarianism and
abundance), rentism (hierarchy and abundance),
socialism (egalitarianism and scarcity), and exterminism
(hierarchy and scarcity). Planning is but one of many facets impinging upon the
extent to which resources are centralised for equitable allocation or
concentrated for hierarchical control and, concomitantly, the extent to which
resources are produced and replenished. Central planning and allocation also
underlie efforts to instantiate and provide basic services.
One of the most widely and frequently discussed ideas
related to full automation is the possibility of a basic income (BI), often
conceived as a universal or unconditional basic income (UBI). A basic income is
the allocation of a nominal sum of money on a recurring basic to individuals
who are not required to provide labour in exchange for this income. Much like
the recent resurgence of socialist thought in mainstream political discourse,
basic income is an old idea that has received considerable recent attention in
correlation with full automation. Since 2015, a veritable cottage industry of
popular press and trade books have outlined the case for basic income (Stern 2016; Van Parijs and Vanderborght 2017;
Bregman 2017; Lowery 2018), with UBI serving
as a cornerstone of Andrew Yang’s 2020 presidential campaign (Yang 2019).
Advocacy for (U)BI tends to stress the increased power workers can enjoy when economic livelihood is not solely tied to wage labour, often stipulated as a necessary condition to address the diminishing need for human labour amidst the rise of full automation. With a UBI in place, workers could potentially choose not to work for certain periods of time, thereby increasing their individual negotiating power and the holistic power of labour (Srnicek and Williams 2015, 120). So long as it is sufficient to provide basic sustenance, allocated unconditionally, and a supplement to welfare programs (rather than a replacement), the working class can experience greater “voluntary flexibility” as opposed to precarity, instability, and insecurity (Srnicek and Williams 2015, 119; 212). At the individual level, UBI could function as a mechanism for providing universal access to resources while guarding against overuse (Frase 2016).
Criticisms
of UBI stress the possibility of UBI as a salve to libertarian and neoliberal
ideologies intent on replacing welfare programs with a lump monetary sum, a
“full marketization of the welfare state” (Bastani 2019, 225). The concern is that pairing
UBI with full automation does not alter the relationship between ownership and
labour and, rather than augmenting the power of labour against ownership, risks
a “miserable penury” for people whose labour potential is seen as useless and
cordoned off from further economic and social mobility (Dyer-Witheford, Kjøsen and Steinhoff 2019, 150-151).
In addition
to Srnicek and Williams insisting on UBI as a
supplement to welfare programmes, Bastani (2019, 215, 217) accounts for UBI as a compliment to
five essential basic services provided on an unconditional basis: housing,
transport, education, healthcare and information, wherein “information” is
understood as “media production and connectivity”. Under Bastani’s
model, the state plays an indispensable role in “procurement with local worker
cooperatives building homes, hospitals and schools as well as performing
catering, maintenance, cleaning, and support services,” indicating a continued
role for private ownership with the caveat “the leverage of anchor institutions
will only expand”, given the role of worker-owned businesses more suited to
address the common good (217).
At this point, we can now more specifically sketch
“digital socialism” as a real utopian project advocating for full automation as
a utopian beacon enjoined with the viable and practical pursuit of the
following:
· Generating
New Ethics, Values, and Arrangements for Labour
o Undoing
the capitalist work ethic and gendered associations with domestic labour
o Shortening
the formal working week
o Pursuing
collectivity as a means of instituting social harmony, reshaping institutional
arrangements, and ensuring collectivist values can be built into automated
technologies
· Centralised
Planning
o Planned
allocation predicated on democratic participation from contributors and
recipients of goods and services
o A
network of localised worker-owned businesses, banks, and financial services
operating under a “municipal protectionism” that guards against capital flight
o Central
federal banking that supports and protects localised businesses and finance
· Basic
Services
o Universal
access to publicly-developed education, transport, housing, healthcare, and
media connectivity and production
o A basic
income apportioned unconditionally as a supplement to universally-available
basic services
From this foundational point, the foremost opportunity in
pursuit of these goals is to ensure ideas about the relationship of full
automation to socialist imperatives do not fall into traps laid by cultural
ideas about the nature of autonomy and technological autonomy as an inevitable
threat to workers.
“Autonomous technology” refers to both a long-standing
cultural fear about the social implications of technological progress and the
functional ability for technologies to operate free from direct human
intervention. In terms of the latter, degrees of automated capacity are often
described in terms of technological autonomy. A fruitful example is
self-driving cars, often referred to as “autonomous” cars. Levels of automation
designated by the Society of Automotive Engineers (SAE) range from “zero
autonomy” to “full autonomy” along a successive scale of automated capability
running from zero to five (with zero as “no automation” and five as “full
automation”) (NHTSA 2019). Achieving full automation
means that, in the case of self-driving cars, the vehicle can perform driving
tasks that do not require a human to operate or intervene (although the
technology may allow for human manual operation). As a cultural fear,
autonomous technology refers to a belief that technology has “gotten out of
control and follows its own course, independent of human direction” (Winner 1977, 13). One of the primary concerns of
functional technological autonomy is the realisation of such fears as
reinforced by pop culture depictions (i.e. the Terminator and Westworld franchises), mainstream news headlines declaring
“The Future Has Lots of Robots, Few Jobs for Humans” (McNeal
2015), and popular press books such as Our
Final Invention: Artificial Intelligence and the End of the Human Era
(2013).
Even as recent scholarly discourse stresses the utopian potential of full automation, other pervasive cultural discourses about the dystopian “hegemonic” or “apocalyptic” implications of technological autonomy advance ideas about the possibilities for widespread social control or “agents of doom” posing an existential threat to human vitality (Nye 2004, 171). Where the cultural fear of autonomous technology is primarily “the question of human autonomy held up to a different light,” (Winner 1977, 43), one of the primary fears about self-driving cars and other present-day autonomous technologies is the direct threat to posed to the autonomy of human labourers.
Books such as The Second Machine Age: Work, Progress, and Prosperity in a Time of Brilliant Technologies (2014) and Rise of the Robots: Technology and the Threat of a Jobless Future (2015) stress the increasing capability of intelligent machines to perform cognitive tasks once believed to be the unique purview of human intellect. Postulations about a forthcoming “automation wave” (Ford 2015) posit an impending tide of machines poised to “steal” the jobs of human labourers and threaten to deepen levels of socio-economic inequality as human workers are displaced – and replaced – in corollary with the escalation of automation towards technological autonomy. The threat to the autonomy of human labour supposed by technological autonomy epitomises many of Marx’s concerns about strife and competition within the working class and the intensification of this circumstance when capitalist production pits machines against humans.
In Capital, Marx describes the labouring capacity of machines as pitted against human workers and, because of this competition: “The self-valorization of capital by means of the machine is related directly to the number of workers whose conditions of exis- tence have been destroyed by it” (Marx 1867, 557). Just as, with respect to commodities, “the devaluation of the world of men is in direct proportion to the increasing value of the world of things” (Marx 1844, 271) [emphasis in original], Marx posits a similar corollary relationship between humans and machines whereby the ability for human workers to seek and procure payment for their labour is invariably diminished the more ownership turns to machinic production. Ever the foresighted critic, Marx recognised that “machinery necessarily throws men out of work in those industries into which it is introduced, it may, despite this, bring about an increase of employment in other industries” (Marx 1867, 570). While technological innovation has historically led to the emergence of new industrial paradigms that reconfigure the types – rather than the amount – of labour required (Bastani 2019), the contemporary narrative that “this time is different” indicates an unprecedented and inalterable risk of permanent displacement due to the humanlike intelligence of increasingly autonomous technology (Brynjolfsson and McAfee 2014; Ford 2015).
Thus,
despite the worthiness of drawing out the utopian potential of full automation
amidst recent debates, such debates neglect the tendency of full automation to
connote fearful notions of autonomous technology and what it portends for the
potential of workers to direct their individual and collective capacities
towards fruitful socio-economic gains. In this vein, foregrounding full
automation risks reinforcing and unintentionally capitulating to Marx’s
concerns about competition between humans and machines and its potential to
agitate intra-class strife among human workers. Applied towards socialist
pursuits, the dichotomy between human and technological autonomy is a critical
gap between the ability to imagine the utopian potential of full automation and
pursue a democratic egalitarianism that can realise strategic imperatives that
make full automation a viable venture.
To mitigate
this gap, I offer human-machine autonomy (Cox 2018) as a
conceptual frame for recognising that autonomy is not couched in a singular
entity, be it human or machine. Within this mindset,
understanding human-machine autonomy as a shared condition among humans and
highly-automated technologies resists misconceptions about autonomy as an
innately dominating force and capitulation to capitalist ideologies around
labour and class composition. The next section unpacks aspects of autonomy amenable
to this line of thinking, before pulling from autonomist Marxism theory to
stress autonomy as a shared condition between humans and technologies and
human-machine alignments against capital.
Ideas about technological and human autonomy as separate
and discreet forces existing in negative correlation arise from illusory
notions of autonomy as the sole province of an individualistic self. Autonomy
is a “political or moral conception that brings together the ideas of freedom
and control” often conceived as the ability to be “self-governing, independent,
not ruled by an external law or force” (Winner 1977, 16).
Scholars of technology and identity, however, reject the conception of an
individual self as the source and purview of autonomous potential, particularly
on the grounds that the autonomy of the self is a politics of domination. In
her “Cyborg Manifesto”, Donna Haraway problematises autonomy as emerging from the relationship
between the self and the other. For Haraway, the self is one who is not dominated, a
non-domination understood only in relation to the dominance of the other. The
supposed ontological nature of the self is “to be autonomous, to be powerful” (Haraway 1990, 219). The ability
to experience freedom and control is therefore tied to the ability to impose
one’s will onto others, a “tragedy of autonomy” that valorises the supremacy of
the self through the domination of the other (Haraway 1990, 219). In their recent critique of
technology as a “surrogate humanity”, Atanasoski and Kalindi (2019, 136)
attempt to further liquidate notions of autonomy in relation to a dominating
self, citing the “myth of the autonomous human” as the product of a “racial
fetish of post-Enlightenment thinking” emanating from colonialist histories
built on subjugation and servitude and attendant notions of autonomy as a
possibility for those who possess mastery and control over the subjugated and
servile.
Such ideas about
autonomy as a dominating self extends to technological autonomy, as the
cultural fear of autonomous technology expresses itself as not only the loss of
control over machines but as “the style of absolute mastery, the despotic,
one-way control of the master over the slave” (Winner
1977, 20). Viewed through this lens, the fear of autonomous technology can
be understood as bound up in the perceived inability for humans to dominate a
technological other and the ability for a technological other to exert the same
type of domination humans pursue through autonomous will.
In other
words, for all that the illusory concept of autonomy as the purview of the self imparts about the politics of domination, one of the
most critical points is that humans and technology share the same root
conception of autonomy. Human and technological autonomy is not a matter of
“here” and “there” but a shared condition inadequate to delineation along lines
of a human or machine and, instead, invokes the same questions around the pursuit
and application of power, freedom, and control. As Haraway
notes, technology is not an object to be “animated,
worshipped and dominated. The machine is us, our processes, an aspect of our
embodiment” (Haraway 1990, 222).
Where autonomy “cannot simply be understood as freedom from others” (Baker and Hesmondhalgh
2013, 40), these “others” include both human and machine counterparts. Moreover, instead of conceiving of autonomy
as self-set life against or apart from an “other,” recognizing that we are “socially constituted by others beyond
themselves” (Baker and Hesmondhalgh
2013, 40) imparts of a sense of how our autonomous potential is truly a
question of our autonomy. In other
words, conceived in opposition to dualistic conceptions, autonomy is always a
shared condition among humans and between humans and machines, even though
autonomy is not equitably afforded or experienced. This does not entail a
deterministic relationship, however, as economic relations, culture, legal
frameworks, and other vectors constitute the circuity that gives human-machine
autonomy its variable charge. Notions of autonomy with respect to capitalist
relations and technology underscore human-machine autonomy as a shared
condition among humans and machines the way capitalism organises and patternises possibilities for autonomy among ownership,
workers, and machines.
Notably, Andrew Feenberg describes
“operational autonomy” as a facet of capitalist ownership that incorporates
autonomous potential into organisation, machinic, and
workflow processes:
Operational autonomy is the power to make strategic choices among alternative rationalizations without regard for externalities, customary practice, workers' preferences, or the impact of decisions on their households. Whatever other goals the capitalist pursues, all viable strategies implemented from his peculiar position in the social system must reproduce his operational autonomy. The ‘metagoal’ of preserving and enlarging autonomy is gradually incorporated into the standard ways of doing things, biasing the solution to every practical problem toward certain typical responses. In industrial societies, strategies of domination consist primarily in embedding these constancies in technical procedures, standards, and artifacts in order to establish a framework in which day-to-day technical activity serves the interests of capital (Feenberg 2002,76).
Understood in this light, capital implants self-serving
notions of autonomy into processes that carry through to the fabric of material
existence so that the autonomous potential of capital is reproduced and
enhanced. To the extent that operational autonomy is a hegemonic imposition of
capital, workers possess a counterhegemonic potential, a “reactive autonomy”
that Feenberg (2002, 84)
otherwise refers to as a “margin of maneuver”. This
reactive autonomy entails the ability of workers to leverage capitalist
technology for the purposes of “controlling work pace, protecting colleagues,
unauthorized productive improvisations, informal rationalizations and
innovations” (Feenberg 2002, 84),
and otherwise countervailing the operational autonomy of ownership. Reactive
autonomy is a margin of manoeuvre because the degree to which workers exercise
autonomy can expand or contract, as can the operational autonomy of capital.
Much like autonomy among humans and technology, operational, and reactive
autonomy are not bracketed off from one another and instead exist as
co-constituted forms of autonomy inflecting upon on one another even though it
is not supposed that reactive autonomy ever exceeds operational autonomy or
reactive autonomy is the exemplar way for workers to attain and experience
autonomy. Automation plays a variable role in this dynamic, as it
increases management's autonomy only at the expense of creating new problems that justify workers' demands for an enlarged margin of maneuver. That margin may be opened to improve the quality of self-directed activity or it may remain closed to optimize control (Feenberg 2002, 96).
To the mind of the capitalist, regardless of the degree
of freedom or control afforded to labour, capitalist exchange “maximizes
autonomy in general, promising liberation of the human essence from fixed
definitions” (Feenberg 2002, 162),
since ongoing acquisition and accumulation are infinite and therefore entail a
range of shifting arrangements that increase both operational and reactive
autonomy in the aggregate.
Of course,
this does not hold up to baseline Marxian scrutiny, as reactive autonomy is an
autonomy conceived and experienced only within the auspices of capitalist
exploitation, alienation, and expropriation of surplus-value, as if operational
autonomy was a natural phenomenon ensconced in some ineffable firmament and not
the result of historical processes predicated on vouchsafing power and control
in the hands of a dominant few. Nonetheless, reactive autonomy reinforces the
central idea of human-machine autonomy: autonomy is a shared condition
experienced with varying intensities relative to critical socio-economic inputs
shaping how autonomy is conceived, pursued, attained, and experienced.
Furthermore, reactive autonomy shines a light on the autonomy of labour,
understood by Autonomist Marxism as not only the autonomous potential of labour
within capital, but the recognition that labour already possesses the ability
to be autonomous from capital.
While the
preceding stresses the autonomy of human-machine autonomy as one resistant to
traditional notions of autonomy as the purview of the dominating self and,
instead, a shared condition among humans and technology shaped by relations to
production and other critical vectors, the final section draws from Autonomist
Marxism to recalibrate a particular strand of Autonomist thought that considers
technology as the means for capitalist domination and the autonomy of workers
as the potential to overcome such technology through class conflict. Rather
than positing capitalist technology as the dominating force, or the force that
must be dominated, the goal in this final section is to reframe this argument
in terms that seek to illuminate how dichotomies of human/technology and
domination/control are apt to reinforce the individualistic drives of
capitalist competition that pit workers against one another and fracture
opportunities for solidified class struggle. Conceiving of autonomy as a shared
condition among humans and machines emphasises the commonality already at hand
among labour and the ability to draw from the shared potential for autonomy to
maximise its potential in work arrangements and the overarching struggle
against capital. By doing so, the
ability to align full automation with strategic imperatives for socialism can
evade technological dystopia, maximise utopian potential, and otherwise resist
capitulation to strife among labour entities (human and machine).
Autonomist Marxism is a branch of Marxian inquiry that
affirms the potential of labour distinct from capitalist arrangements (Negri 2005; Berardi
2009; Tronti 1966/2019),
foregrounding labour’s “creative human energy” and the labourer as the “active
subject of production, the wellspring of the skills, innovation and cooperation
on which capital depends” (Dyer-Witheford 1999, 65). Autonomy, in this context, also refers
to “labor’s fundamental otherness from capital and also the recognition of variety within labor” (Dyer-Witheford 1999, 68). The variety within labour speaks
to the recognition that capitalist labour is not a uniform series of functions
and workers can strive for circumstances best suited to differentiated skills,
innovation, and cooperation, even as wage labour imposes itself as in
restrictive force for worker autonomy. Labour’s “otherness” from capital, on
the other hand, recognises the ability of the working class to exist apart from
capitalism, while capitalism cannot exist without the working class. Since capitalism can only instantiate and
maintain power through the institutionalisation of its aims, the autonomous
potential of labour lies in a “non-institutionalized political power” unique to
the working class (Tronti 1966/2019,
247), whose position as the subject of production entails an innate
power unbeholden to institutional forms or the
auspices of capitalist accumulation.
The
Autonomist position on technology tends to correspond to the notion of autonomy
as the grounds for domination. On one hand, technology is the means for capital
to control and dominate workers while, on the other hand, capitalist technology
is the thing that should be dominated, as through class conflict workers can
upend capitalist technology and subsequently remake it in the image of
socialist ends, leveraging the ability for workers’ autonomous potential to
break from capital and harness their “invention power” (Dyer-Witheford 1999, 69-71).
From this perspective, Autonomist views of digital technologies reinforce the
way emerging technologies are developed and deployed to be amenable to
capitalist relations. In his description of the emerging “cognitariat”,
Berardi (2009, 35) cites digital technologies and
network connectivity as giving rise to the ubiquity of cognitive labour
performed without deference to formal work arrangements or social existence, a
“creation of technical and linguistic interfaces ensuring the fluidity both of
the productive process and of social communication”. Matteo Pasquinelli’s
read on information technology entails a similar transformative process, with
regard to Marx’s organic composition of capital: “living information is
understood as continuously produced by workers to be turned into dead
information crystallized into machinery and the whole bureaucratic
apparatus of the factory” (Pasquinelli
2015, 55).
In both cases, digital technology is conceived as a tool to exacerbate
capitalist domination. Therefore, it should be surmounted by working class
revolution. I do not necessarily quibble or find fault with these assessments
but, rather, point out the opportunity for the Autonomist perspective to apply
its foundational spirit to digital technologies and recognise human-machine
autonomy as an opportunity to consider political revolution as a coalition of
human and technological workers based on their common subjectivity as labouring
entities and the recognition for the mutual endeavours of humans and automated
technologies to help realise the aims of digital socialism. If the original aim
of Autonomist thought was to foreground the autonomy of workers as an inherent
feature of class struggle, extending Autonomist thought outwards towards
potential socialist futures demands consideration of another possibility: full
automation need not be a choice between a tool to shed capitalist dominion or
an inert infrastructure awaiting a political revolution to rewire its
programming. Instead, a third option emerges: automated technologies as
co-constituted with human workers and the working class. By understanding
autonomy and autonomous production as a shared condition, human-machine
autonomy can frame the struggle against capital as a form of solidarity among
autonomous production undertaken across lines of human and technological
performance based on their shared position against capital. This position is
best illuminated through Mario Tronti’s problematisation of a working-class ideology and the
“strategy of refusal.”
Tronti describes the unnecessity of developing
an ideology unique to the working class, since the working class is a “a
reality antagonistic to the entire system of capitalism”, an ontological
position that means workers exist irrespective to capitalism and are not
inevitably bound to circumstances enabling the development and persistence of
capitalist exchange (Tronti 1966/2019,
6). The working class possesses the potential to exist beyond
capitalism, whereas capitalism cannot exist without the exploitation of the
worker. Should the working class accept the necessity of ideology, their
struggle would become a “passive articulation
of capitalist development” (7) [emphasis in
original]. If the working class needs no ideology,
and the pure fact of their autonomous production is sufficient, they are allied
with machines as non-ideological and autonomous workers freighted with
ideological dimensions by capital. Rather than conceive of technologies as
allied with capital by virtue of their operational deployment against workers,
we should recognise that ideology is neither a necessity for workers nor
machines and both are subject to the imposition of capitalist ideologies with
respect to the ways work is arranged and carried out.
Humans and
automated technologies are both programmed to perform computational tasks
carried out in accordance with the imperatives encoded into such programming (Bucher 2018). Computers are programmed via the input of
computer code that dictates how to operate, just as human behaviour is directed
by technological and social codes that impart ideas about how humans should
operate. In the context of work, any worker striving to build the latest
iteration of AI or leverage AI in formal work arrangements occupies an allied
subjectivity with technology insofar as both are inflected with ideologies
about capitalist work in spite of the shared unnecessity for ideology or work
to be undertaken in accordance with capitalist principles. To attempt to break
from capitalist technologies is to break from entities allied with workers, as
this break is to affirm ideologies about the ontological existence of
technologies and their “passive” position within capitalist orders. Further, to
break from technology is to revert back towards the ideology of autonomy that
insists upon domination as the means for political freedom. Extending Tronti’s “strategy of refusal” offers a means to refuse
ideological assumptions about autonomy and the split between the autonomous
productivity of humans and machines.
The strategy of refusal acknowledges the ability for the
working class to halt capitalist production by refusing to carry out capitalist
demands or undertakings. Understood as both “the
refusal to collaborate actively in capitalist development, [and] the refusal to
put forward a positive programme of demands” (Tronti 1966/2019, 255), this strategy spotlights
the autonomy of the working class to exist apart from capitalism and therefore
use collective labour power as a means to advance the power of labour. Power,
in this context, is the political power to recognise the autonomous potential
to refuse capital and the power to cease productive activities that accord
economic and social power to capital (Tronti
1966/2019, 256). By expanding the political valence of this refusal
to consciously incorporate technological counterparts, the working class aligns
all possible autonomous production as part of its refusal strategy and thwarts
ideological ideas about autonomy as a source for domination and control. Solidarity
with automated technologies, then, is not only possible; it is critical as a
means of evading dystopian conceptions of technological autonomy, resisting
ideological assumptions about autonomy, and undertaking political praxis geared
towards maximising worker autonomy within capitalist as a means to move beyond
its horizons.
While the preceding offers a foundation for typifying
digital socialism and incorporating a human-machine autonomy that stresses the
shared conditions humans and machines occupy with respect to capital,
opportunities abound from this foundation. Scholars should consider
relationships between eco-socialism (Pepper 2002; Huan 2014) and full automation,
especially potential oppositions between raw resources necessary to develop
such technologies and the environmental consequences of continued technological
development. Additionally, while the politics of full automation largely
corresponds to postcapitalist perspectives,
Blockchain advocates imagine Blockchain automation as a source for Libertarian
autonomy conceived as liberation from central banking and the state (Greenfield 2017; Swartz 2017),
indicating a critical need to consider the Blockchain’s
decentralised structure and politics with an eye towards socialist imperatives.
To
stress the utopian and joyous potential of digital socialism, human-machine
autonomy should also be used to expand ideas around “acid communism” (Gilbert 2017; Fisher 2018). Acid communism is a “provocation and a
promise” (Fisher 2018, 757) oriented towards
recapturing the joyous spirit and harmonious possibilities of countercultural
politics and lifestyle. Where neoliberalism established itself as a sensible
form of individualism defined in contrast to ideas of collectivity and communal
living emerging out of the 1960s counterculture, acid communism urges
re-establishing countercultural pursuits for “the convergence of class
consciousness, socialist-feminist consciousness raising and psychedelic
consciousness, the fusion of new social movements with a communist project, an
unprecedented aestheticisation of everyday life” (Fisher
2018, 758). It is, in other words, a recuperation and continuation of a
cultural project otherwise stripped of its revolutionary potential and grouped
into a libertarian ethos underpinning the emergence and global expanse of
Silicon Valley (Turner 2010). Human-machine autonomy,
then, can stress the collective reservoirs of autonomy already at hand for a
collective consciousness that simultaneously seeks to stand down capitalist
power imposition and uplift the ability to live a joyful life of meaningful
pursuit indicative of Marxian aims for a worker’s paradise.
Atanasoski, Neta and Kalindi Vora. 2019. Surrogate
Humanity: Race, Robots, and the Politics of Technological Futures. Durham,
NC: Duke University Press.
Baker,
Sarah and David Hesmondhalgh.
2013. Creative Labour: Media Work in
Three Cultural Industries. London: Routledge.
Barbrook, Richard and Andy Cameron.
1996. The Californian Ideology. Science
as Culture 6 (1): 44-72.
Barrat, James. 2013. Our
Final Invention: Artificial Intelligence and the End of the Human Era.
London: Macmillan.
Bastani, Aaron. 2019. Fully Automated Luxury Communism.
London: Verso Books.
Berardi, Franco. 2009. The Soul at Work: From Alienation to Autonomy. Los Angeles, CA: Semiotext (e).
Bloch, Ernst. 1918/2000. The Spirit of Utopia. Redwood City, CA:
Stanford University Press.
Bostrom, Nick. 2014. Superintelligence: Paths, Dangers,
Strategies. Oxford: Oxford University Press.
Bregman, Rutger.
2017. Utopia for Realists: And How
We Can Get There. London: Bloomsbury Publishing.
Brynjolfsson, Eric and Andrew McAfee. 2014. The Second Machine Age: Work, Progress, and
Prosperity in a Time of Brilliant Technologies. New York City: WW Norton
& Company.
Bucher, Taina. 2018. If... Then: Algorithmic Power and Politics.
Oxford: Oxford University Press.
Chomsky, Noam. 2016. Interview – Noam
Chomsky on Anarchism, Communism and Revolutions. Interview by C.J. Polychroniou. Truthout. Accessed Nov. 15, 2019. https://truthout.org/articles/noam-chomsky-on-anarchism-communism-and-revolutions/
Cox, Christopher. 2018. Autonomous Exchanges: Human-Machine Autonomy
in the Automated Media economy. Accessed Oct. 26, 2019. https://scholarworks.gsu.edu/cgi/viewcontent.cgi?article=1003&context=fmt_dissertations
Danner, Chas. 2015. Jeremy Corbyn and the British Left’s Rebellion. Intelligencer.
Accessed Oct. 22, 2019. http://nymag.com/intelligencer/2015/09/jeremy-corbyn-and-the-british-lefts-rebellion.html
Dyer-Witheford, Nick. 1999. Cyber-Marx:
Cycles and Circuits of Struggle in High-Technology Capitalism.
Urbana, IL: University of Illinois Press.
Dyer-Witheford, Nick, Atle Mikkole Kjøsen
and James Steinhoff. 2019. Inhuman Power:
Artificial Intelligence and the Future of Capitalism. London: Pluto Press.
Feenberg, Andrew. 2002. Transforming Technology: A Critical Theory
Revisited. Oxford: Oxford University Press.
Frase, Peter. 2016. Four Futures: Life After Capitalism. New
York City: Verso Books.
Frizell, Sam. 2015. Here’s How Bernie Sanders
Explained Democratic Socialism. Time. Accessed Oct. 22, 2019. https://time.com/4121126/bernie-sanders-democratic-socialism/.
Gilbert, Jeremy. 2017. Why the time has
come for ‘Acid Corbynism.’ The
New Statesman. Accessed Nov. 7, 2019. https://www.newstatesman.com/politics/uk/2017/10/why-time-has-comeacid-corbynism
Haraway, Donna. 2016. Staying with the Trouble: Making Kin in the Chthulucene. Durham, NC: Duke University Press.
Haraway, Donna. 1990. A Manifesto
for Cyborgs: Science, Technology, and Socialist Feminism in the 1980s. In Feminism/Postmodernism, ed. Linda J.
Nicholson, 190-233. New York: Routledge.
Hester, Helen. 2018. Xenofeminism. Hoboken:
John Wiley & Sons.
Huan, Qingzhi.
2014. Eco-Socialism as Politics.
New York City: Springer.
Kember, Sarah.
2003. Cyberfeminism and Artificial
Life. London: Routledge.
Lowery, Annie. 2018. Give People Money: How a Universal Basic
Income Would End Poverty, Revolutionize Work, and Remake the World. New
York City: Crown Publishing.
Marx, Karl. 1867. Capital Volume One. London: Penguin.
Matthews, Dylan. 2016. Inside Jacobin. Vox. Accessed Oct. 16, 2019. https://www.vox.com/2016/3/21/11265092/jacobin-bhaskar-sunkara
McNeal,
Marguerite. 2015. Rise of the Machines: The Future has Lots of Robots, Few Jobs
for Humans. Accessed Oct. 27, 2019. https://www.wired.com/brandlab/2015/04/rise-machines-future-lots-robots-jobs-humans/
National
Highway Traffic Safety Administration. 2019. Automated Vehicles for Safety.
Accessed Oct. 27, 2019. https://www.nhtsa.gov/technology-innovation/automated-vehicles-safety
Negri, Antonio. 2005. The Politics of Subversion: A Manifesto for
the Twenty First Century. Cambridge: Polity Press.
Nye, David.
2004. Technological Prediction: A Promethean Problem. In Technological
Visions: The Hopes and Fears that Shape New Technologies, eds. Marita Sturken, Douglas Thomas,
and Sandra Ball-Rokeach, 159-176. Philadelphia:
Temple University Press.
Paden,
Roger. 2002. Marx's Critique of the Utopian Socialists. Utopian Studies
13 (2): 67-91.
Pasquinelli, Matteo.
2015. Italian Operaismo and the Information Machine. Theory, Culture & Society 32 (3): 49-68.
Pepper, David. 2002. Eco-Socialism:
From Deep Ecology to Social Justice. London: Routledge.
Phillips, Leigh and Michael Rozworski. 2019. The People's Republic of Wal-Mart: How the
World's Biggest Corporations Are Laying the Foundation for Socialism.
London: Verso.
Schwartz, Joseph M. 2017. A History of
Democratic Socialists of America 1971-2017. Accessed Oct. 16, 2019. https://www.dsausa.org/about-us/history/
Sperber, Nathan. 2019. The Many
Lives of State Capitalism: From Classical Marxism to Free-Market
Advocacy. History of the Human Sciences 32 (3): 100-124.
Srnicek, Nick
and Alex Williams. 2015. Inventing the Future: Postcapitalism
and a World Without Work. New York
City: Verso Books.
Stiegler, Bernard. 2018. Automatic
Society: The Future of Work. Hoboken: John Wiley & Sons.
Sunkara, Bhaskar.
2019. The Socialist Manifesto: The Case for Radical Politics in an Era
of Extreme Inequality. New York City: Hachette UK.
Swaim, Barton. 2019. Political Books: The Rise of
Equality Chic. Wall Street Journal,
April 26. Accessed Oct. 22, 2019. https://www.wsj.com/articles/political-books-equality-chic-11556290466
Tronti, Mario. 1966/2019. Workers
and Capital. London: Verso.
Van
Dijck,
José. 2013. The Culture of Connectivity:
A Critical History of Social Media. Oxford: Oxford University Press.
Van Parijs, Phillippe and Yannick Vanderborght.
2017. Basic Income: A Radical
Proposal for a Free Society and a Sane Economy. Cambridge, MA: Harvard
University Press.
Weissman,
Jordan. 2019. Should the U.S. Trash Capitalism? Slate. Accessed Oct. 22, 2019. https://slate.com/business/2019/05/socialist-manifesto-bhaskar-sunkara-socialism-bernie-sanders.html.
Wright, Erik Olin. 2010. Envisioning Real Utopias.
London: Verso Books.
Yang,
Wesley. 2019. What Andrew Yang
Means. Washington Post. Accessed Oct. 26, 2019. https://www.washingtonpost.com/outlook/2019/10/25/what-andrew-yang-means/?arc404=true
Christopher M. Cox
Christopher M. Cox, PhD is Assistant Professor of Communication and Converged Media at the University of Akron. Researching in the areas of digital media theory, political economy, and cultural studies, his work examines the industrial development of digital media platforms and automated media technologies.