2 WHAT IS A MACHINE
✅ What Is a Machine? — Core Concepts
1. Machines Everywhere
Bryant defines machines very broadly. They are not limited to technology or human-made tools.
Examples:
• Trees, quartz crystals, recipes, novels, frogs, cells, tanuki, books.
• Machines don’t need design, purpose, or users to count as machines.
A world is made of machines.
This term avoids the baggage of “object,” which assumes:
• attachment to a subject who perceives it,
• definition by properties and predication.
2. From Objects to Operations
Traditional metaphysics treats things as objects with:
• a subject,
• a set of qualities or predicates,
• an unchanging essence beneath changing accidents.
But Bryant shifts the focus:
The question is not “What qualities does something have?” but “What does it do?”
A machine = something that performs operations.
3. Operations, Inputs, Outputs
Drawing on Ian Bogost, an operation:
• takes one or more inputs,
• transforms them,
• produces an output.
Inputs can arise:
• from outside (e.g., a flower draws water from the soil),
• from inside (e.g., cells releasing chemicals to other cells).
A machine is:
a system of operations transforming inputs into outputs.
4. Examples Reimagined as Machines
Rather than seeing entities as static things with properties, Bryant reinterprets them dynamically:
• Tree: transforms water, light, nutrients into wood, oxygen, seeds, etc.
• Recipe: a machine acting on cooks, ingredients, utensils.
• Book: a machine affecting language, readers, characters, institutions.
• Scientific paper: a machine that proposes observations, constructs devices, generates
experiments.
• Frog: a machine catching insects, producing songs, navigating water, emitting CO₂,
reproducing.
5. Production vs Representation
Machines do not express or represent; they produce.
This opposes the idea that:
• books express meaning,
• the unconscious represents symbols,
• objects mirror human perception.
Instead:
• Machines are productive factories.
• The world is a fabric of ongoing processes and outputs.
6. Deleuze & Guattari: Factory vs Theater
They argue Freud made a critical error:
• Freud discovered the unconscious as productive (factory),
• but then reduced it to representational expression (theater) through Oedipus.
Bryant applies the same contrast to all machines:
• Factory = production, operation, transformation.
• Theater = expression, representation, meaning.
Machines are always factories, not theaters.
✅ Big Takeaway
A machine is not defined by what it is but by what it does.
• It performs operations.
• It takes inputs and produces outputs.
• It is material and active.
• It is not reducible to meaning, expression, or representation.
This becomes the foundation for his ontology:
Being = machines all the way down.
If you'd like, I can turn this into a visual diagram, concept map, or bullet-point guide for
reference—just say the word!
✅ 1. Machines Have Two Dimensions
Bryant argues that every machine has two distinct but inseparable aspects:
a) Virtual Proper Being (Powers)
• What a machine can do — its capacities or operations.
• These powers exist even when not exercised.
• Example:
◦ Iron can rust, even if it never does in outer space.
◦ A match has the power to burn, even if unlit.
b) Local Manifestation (Products/Outputs)
• What a machine actually does when its powers are exercised.
• These are the concrete results of operations: rust on iron, a flame from a match, oxygen
from a plant.
✅ 2. Powers Are Real, Even When Inactive
Drawing from George Molnar, Bryant explains five key features of powers:
1. Directedness
◦ Powers aim at specific kinds of effects (e.g., photosynthesis → oxygen).
2. Virtuality & Range
◦ Powers can produce more manifestations than appear at any given moment.
3. Independence from Manifestation
◦ A power exists even when not being exercised (e.g., a skunk’s spray).
4. Actuality of Powers
◦ Powers are real features of things, not hypothetical or mental.
5. Intrinsic & Objective
◦ Powers belong to the machine itself, not to observers or contexts.
Terminology distinction:
• Power = the capacity itself.
• Operation = when a power is exercised.
✅ 3. Manifestations Are Products, Not Perceptions
Manifestations are not for someone — they simply happen:
Types of manifestations:
1. Qualitative (e.g., a tan, color change, brittleness in steel).
2. Agentive (changes in behavior or activity, like shivering or love).
3. Material (outputs that leave the machine — oxygen from trees, a diploma from an
educational system, speech).
Crucially:
• Qualities (like color, temperature, texture) are not inherent properties but effects of
operations.
✅ 4. Inputs Trigger Manifestations
Operations can be triggered:
• Externally (sunlight, soil nutrients, wind, heat).
• Internally (radioactive decay, sequential thoughts).
A manifestation = operation of a power on an input.
✅ 5. Local Manifestations Vary With Conditions
• Machines do not express fixed properties — they produce effects depending on context.
• Example: The same grape stock produces different wines in different regions.
Machines form relational ecologies, interacting and mediating each other.
✅ 6. Powers Are Plastic and Change Over Time
Machines:
• Can gain or lose powers.
• Experience waxing and waning of capacities (e.g., hunger affecting cognition).
• Can transform through learning, tools, encounters, or internal change.
Examples:
• The brain of a pianist differs from that of a mechanic.
• A blind person with a cane has different cognitive powers than without.
• A caterpillar becomes a butterfly.
• Steel loses malleability in extreme cold.
Plasticity is always present, though not infinite—each machine has limits proper to its kind.
✅ 7. We Discover Powers Experimentally
Echoing Spinoza:
“We do not know what a machine can do.”
The being of a machine is found through:
• Acting on it,
• Altering its relations,
• Observing new manifestations and transformations.
✅ Core Takeaway
A machine is defined not by what it is but by:
• What it can do (virtual proper being — powers),
• And what it actually does (local manifestation — products).
Being is operational, plastic, and context-dependent—not a static set of inherent properties.
✅ 1. Machines Are Binary Because They Operate Through
Couplings
• A machine never operates in isolation—it always interacts with other machines that
supply inputs (flows).
• These coupled interactions produce local manifestations (qualities, activities, outputs).
• Example: A tree is coupled with soil, sunlight, rainfall, animals, and microorganisms.
However, the authors critique Deleuze & Guattari’s claim that machines are always coupled:
• Couplings must be variable, not fixed.
• It’s the presence or absence of a relation that produces different outcomes.
• Example: An infant isn’t always coupled to a breast; sometimes it is, sometimes it isn’t—
this variation is what matters.
Key point:
Relations (couplings) are external to machines—they can be formed, broken, or changed. This
creates different becomings and manifestations.
✅ 2. Trans-Corporeality: Bodies Are Intermeshed and Co-
Affecting
Drawing on Stacy Alaimo:
• Bodies (human and nonhuman) are entangled with their environments.
• Inputs (air, chemicals, food, toxins, waste) come from other machines and shape local
manifestations.
• Example: Pollution enters water via dumps, enters wildlife, then enters our bodies—we
never really throw matter “away.”
Trans-corporeality =
Machines exist in a web of material exchanges that alter their states.
✅ 3. Flows Are Not Passive—They Are Machines Too
• Inputs (flows) are not raw, formless matter.
• Every flow is another machine, with its own powers and operations.
• Example: Nutrients, omega-3 fatty acids, or marble being sculpted each exert their own
influence.
This leads to reciprocal determination:
• Machines modify the flows they act on,
• and the flows modify them in return.
✅ 4. Degrees of Plasticity: Machines Change to Different
Extents
Machines exist on a spectrum from rigid to flexible:
Rigid machines
• Barely altered by inputs.
• Examples: mathematical formulas, bureaucracies, cookie cutters.
Highly plastic machines
• Strongly altered by inputs.
• Examples: organic bodies, brains, political movements, artworks.
No machine is fully rigid or fully plastic—each exists on a continuum.
✅ 5. Great Works of Art as Pluripotent Machines
Artworks are unique kinds of machines:
• They maintain a stable form (a novel, painting, sculpture),
• But are highly plastic in meaning and effect.
They produce different manifestations depending on:
• historical context,
• culture,
• interpretation.
Example: Kafka’s The Trial and The Castle can be:
• political critiques,
• theological texts,
• psychoanalytic caseworks,
• anti-bureaucratic parables,
• existential philosophy, etc.
This is called resonance:
• The artwork affects and is affected by its milieu.
• The milieu actualizes different potentials in the work.
✅ Core Takeaways
✔ Machines manifest differently depending on their couplings and inputs.
✔ Relations are external, variable, and transformative.
✔ Bodies are trans-corporeal, intertwined with other machines and environments.
✔ Inputs are active machines, not passive matter.
✔ Machines range from rigid to plastic, with artworks being especially pluripotent.
✔ Reciprocal influence—machines reshape and are reshaped by what flows through them.
✅ 1. Machines Operate
• Everything is a machine
Trees, recipes, novels, cells, crystals, papers, frogs—machines aren’t just tools or
technologies.
• Not defined by purpose or perception
Machines don’t need a user, designer, or observer to exist.
• Shift from what it is to what it does
Focus on operations, not properties or essences.
• Operation = input → transformation → output
Inputs may come from outside or inside the machine.
• Examples as machines
◦ Tree: transforms water, CO₂, and light into oxygen, wood, seeds.
◦ Book: affects language, readers, institutions.
◦ Recipe: acts on cooks, ingredients, tools.
◦ Frog: catches insects, sings, reproduces, processes air.
• Production over representation
Machines don’t “express” or “mirror” meaning—they produce effects.
• Factory vs theater
◦ Factory = transformation and output (the real nature of machines).
◦ Theater = interpretation and expression (a misleading human frame).
✅ 2. Machines Are Split Between Their Powers and
Products
• Two dimensions of every machine:
◦ Virtual Proper Being (Powers):
Capacities that exist whether used or not.
◦ Local Manifestation (Products):
What appears when powers are exercised.
• Powers are real even when dormant
A match can burn even if unlit; iron can rust even if never exposed.
• Manifestations are effects, not fixed traits
Color, texture, behavior, outputs—these are products of operations, not inherent
properties.
• Inputs trigger operations
External (sun, nutrients, wind) or internal (chemistry, decay, thought).
• Context-dependent results
The same machine acts differently in different relations (soil → grape → wine variation).
• Plasticity of powers
Powers change, grow, weaken, or emerge through:
◦ tools, training, environments, encounters, internal change.
• We learn powers through experimentation
“We do not know what a machine can do” until it’s acted upon.
✅ 3. Machines Are Binary Machines: Trans-Corporeality
• Binary = machines operate through couplings
No machine acts entirely alone—operations occur through interacting with other
machines.
• Relations are variable and external
Couplings can form, break, or shift—this variability shapes outputs and metamorphosis.
• Trans-corporeality
Bodies (human and nonhuman) exchange materials, forces, and effects with their
environments.
• Flows are machines too
Inputs aren’t passive matter—nutrients, toxins, light, marble, air are active machines.
• Reciprocal transformation
Machines alter what they engage with, and are altered in turn.
• Range of rigidity vs plasticity
◦ Rigid machines: little changed by inputs (bureaucracies, formulas, cookie
cutters).
◦ Plastic machines: highly altered by inputs (bodies, artworks, movements).
• Art as pluripotent machine
Stable form + variable manifestation across:
culture, era, interpretation, context.
• Resonance
A machine’s environment activates different potentials within it.