"Imagine you're holding a pen in your hand. How many objects do you have in the palm of your hand?"


"I unscrew the cap from the body of the pen. Now how many objects?"


"Then I pull out the ink reservoir from the body. Now?"


"Suppose I break the plastic body into many pieces. And now?"

Quite a few – however many pieces

Broken pen

Pieces of a whole? Credit: Author, Midjourney

"Then, I get a tube of glue and meticulously stick the pieces back together. So tell me, how many objects do you have now?"


"But the glue isn't dry yet..!"

And so we continue.

In our everyday speech, we don't bother with such distinctions. We see a pen as a singular, functional object; and we name it as such. But when we shift to a metaphysical lens, the boundaries blur. When does a collection of small units of matter become a composite object?

This sets the scene for the puzzle of the Special Composition Question, or the "SCQ", originally proposed by philosopher Peter van Inwagen.

Special Composition Question (SCQ): When, if ever, do objects compose?

The SCQ is a question in mereology, a part of philosophy which is all about parthood – the relations between parts and wholes.

The SCQ has sparked a enormous debate, with philosophers proposing a wild range of answers. We're going to explore some of them.

Answer 1: Sometimes

If the answer to the SCQ is "sometimes", then certain, but not all, combinations of mereological simples give rise to composite objects.

A mereological simple (or "mereological atom") is an object that has no proper parts, or in other words, it is an object that cannot be divided into smaller components.

These composites, in turn, might assemble to create further composite entities. There are a wide array of sometimes views, which I use as an informal umbrella term.

The Ship of Theseus

A composite object made of composite objects? Credit: Author, Midjourney

The challenge with sometimes views is specifying the precise conditions under which composition transpires. What magic formula allows certain combinations of simples or composites to form a new composite, while others remain a mere jumble of mereological simples?

Let's look at a simple options, to illustrate.

The Contact View: Objects compose another only when they're in physical contact. Just a little interrogation makes this view crumble. Does a handshake create a new entity? Are atoms objects, composed of protons and electrons despite not being in direct contact? Is the sea an object? The view is unviable.

The Fastening View: Composition occurs when objects are fastened together – say, with screws, glue, or the like. While it's a prima facie intuitive assertion, its limitations become evident quickly. If we superglue two people's hands together, are they now a composite object? What if the glue is still drying? Considering the atomic structure where particles aren't physically 'fastened,' does it imply the nonexistence of atoms?

We might find this vagueness to be troubling. I certainly do. Determining the exact moment a collection of objects morphs into a single composite object like catching smoke. This lack of clear boundaries complicates questions of identity too, as demonstrated by the enduring Ship of Theseus paradox, which I wrote about here.

Some technical stuff: x is a proper part of O iff it is part of O but not identical with O. Where ‘R' is a relation denoting being–a–proper–part–of:

Proper Parthood ≜ ∃x∃y(Rxy → ¬x=y)

Answer 2: Never (Mereological Nihilism)

Mereological nihilism: objects never compose.

On this view, there are no composite objects. Only mereological simples exist.

Your pen, the stars and everyone you know are just congregations of simples arranged in a particular way.

There's something appealingly elegant about its reductionist stance. It sidesteps the quagmire of vagueness, indeterminacy, and ambiguity that plagues sometimes views.

If we take nihilism at face value, it means accepting that entities we interact with daily – chairs, trees, people – are not 'real' objects.

Nihilists, then, bear the burden of providing an alternative account of how we navigate and understand the world.

The likes of Trenton Merricks and Peter Unger have fleshed out defences of nihilism. Merricks, for instance, advocates for a version of nihilism where, while composites don't exist, the simples arranged 'chair-wise' or 'pen-wise' are real, and can fulfill the roles we typically attribute to composite objects.

There are problems, though. The nihilist needs to do some work to explain emergent properties, like waves in water (I've levelled a response to this criticism here). It seems water can have properties that its constituent parts do not. There are also some inconsistencies with widely accepted theories in physics, such as quantum entanglement and field theory, which I won't get into here.

Some technical stuff: For all objects x and y, if x is a proper part of y then x is identical to y. Where ‘R' is a relation denoting being–a–proper–part–of:

Mereological Nihilism ≜ ∃x∃y(Rxy → ¬x=y)

Answer 3: Always (Mereological Universalism)

Mereological universalism is an answer to the SCQ that sits on the opposite end of the spectrum to mereological nihilism.

Mereological universalism: objects always compose.

Any motley collection of objects, no matter how disparate or seemingly unrelated, forms a composite object.

So, universalism bypasses the need for specific composition conditions.

Historically, this view has been defended by philosophers such as David Lewis and Theodore Sider. They both argued that mereological universalim, though counterintuitive, offers theoretical simplicity and elegance. After all, why insist on carving out specific circumstances for composition when we can posit a universal rule?

By endorsing a limitless proliferation of composite objects, of course, we commit ourselves to a parade of bizarre composite objects. There's an object composed of the Moon and a sixpence (an example I borrow from James van Cleve). There's an object composed of your left slipper, the Eiffel Tower, and a solitary grain of sand on Mars.

The Moon at the Sixpence

A composite object? Credit: Author, Midjourney

But if everything and anything forms a composite object, it begins to feel that the term 'composite object' loses its discerning power. In other words, it no longer distinguishes between meaningful aggregations and arbitrary groups of matter.

Universalism does seem to solve some problems the nihilst had – and we can make sense of universalist's metaphysical greed by suggesting that the way we categorise and distinguish objects isn’t necessarily bound by metaphysics but can be, in part, dictated by our language and conceptual framework. We can carve up the world with our terms and concepts, and guarantee that there will be a composite object that will act as a truthmaker for any true proposition about the world. That's pretty neat, on the face of it.

There's no denying, though, that the universalist implies the existence of some pretty bizarre composites, and also needs to defend their stance from some worries from physicists, including – again – quantum entanglement. More on that later.

Some technical stuff: For all objects x and y, if x and y are both parts of z, then there exists a further object w such that both x and y are parts of w. Where ‘R' is a relation denoting being–a–proper–part–of:

Mereological Universalism ≜ ∀x∀y∃z(Rxz ∧ Ryz → ∃w(Rxw ∧ Ryw))

Answer 4: Another sometimes view – organicism

Peter van Inwagen's answer to his own question is organicism.

Organicism: Composition occurs only if the object constitutes a life

Deer on a hill

Six enormous composite objects in a sea of mereological atoms? Credit: Author, Midjourney

For instance, the molecules constituting a deer form a composite object, but the parts of a non-living entity – say, a watch or a spade – do not.

It's meant to align with our everyday intuitions that we, and other living things, exist as distinct entities, not mere clusters of fundamental particles. Organicism acknowledges this intuition. It reinforces the common sense notion that living things are meaningful wholes that are more than the sum of their parts. It's meant to provide satisfying truthmakers for these beliefs, unlike the sweeping reductionism of the nihilist, who has to do much more work.

But the precise delineation of 'life' is a minefield of its own. I can't help but feel that organicism feels more like a convenience which reduces the question of composition into a question of what we would most like to find truthmakers for.

Some technical stuff: If x and y are proper parts of z and z is organically unified, there exists a further object w that is organically unified and of which x and y are proper parts. Where ‘R' is a relation denoting being–a–proper–part–of and 'O' is a relation denoting being-organicially-unified:

Mereological Organicism ≜ ∀x∀y∀z∀w((Rxz ∧ Ryz ∧ Oxz ∧ Oyz) → ∃w(Oxw ∧ Oyw ∧ Rwx ∧ Rwy))

Answer 5: There's no fact of the matter. Brutal composition.

Brutal composition: There is no definitive answer as to when objects compose.

In other words, there isn't an underlying principle or set of conditions that determines when objects compose or not.

Philosophers like Ted Sider argue that the existence of composite objects is a "brute fact" – something that just is sometimes the case, with no deeper explanation or justification. So, according to this view, it's possible that some collections of objects form a composite, while others do not, and there is no further fact about these collections that explains why they do or do not compose.

Brutal composition implies that it's impossible to predict or articulate when composition occurs.

If you have a high tolerance for unresolved metaphysical facts, you might be able to make peace with brutal composition, but for me, it is more of a signal that something is wrong with the idea of composition altogether.

Is something wrong with the SCQ? A final option: Existence Monism

We haven't arrived at a satisfying conclusion. By now, you may be wondering whether there is something wrong with the SCQ — "When, if ever, do objects compose?". Perhaps it is misdirected.

It starts from an assumption of many-ness, of a reality teeming with separate objects. It then seeks to find out how and when these objects come together to form composite structures.

As an existence monist, I see this as a misstep.

Existence Monism: There is exactly one concrete object — the universe

On this view, all seemingly separate entities we perceive — pens, stars, humans — are simply subregions of the one-and-only object (the universe). They are not truly distinct or separate but merely patterns of order within the unified whole.

So the very concept of objects composing becomes moot.

Of course, existence monism goes against the grain of our usual perception of a world filled with distinct entities. It demands a reevaluation of our language and thought structures, which hinge on identifying and differentiating separate things.

Despite these challenges, existence monism aligns far better with certain interpretations of quantum physics, which propose a fundamentally interconnected reality.

I wrote much more on this subject elsewhere:

Some technical stuff: A concretum x exists, and should any concrutum exist, it must be identical to x. Where "C' denotes being a concretum:

Existence Monism ≜ ∃x(Cx ∧ ∀y(Cy → x=y))

Further reading/watching

Video: The Special Composition Question – Jonathan Tallant – My metaphysics professor from the University of Nottingham explains the SCQ in this great 17-minute video. Very much unlike me, he really knows what he's talking about and is in the thick of the debate today.

Introductory book: Metaphysics: An Introduction – Jonathan Tallant – An accessible introduction to a range of topics in metaphysics including material constitution, told through the lens of truthmaking. Available from Wordery

Advanced book: Material Beings – Peter van Inwagen – A seminal text where he introduces both the SCQ and organicism. A challenging read for newcomers but an important and precise resource. Available from Blackwell.

Advanced paper: The Challenge to Nihilism – Harold Noonan – A critical dissection of modern debate around nihilism. It's readable, and sets the scene of the debate well. Noonan – another professor at my alma mater – challenges the assertion that nihilists cannot satisfactorily answer the logically analogous question to the SCQ, termed the 'Special Arrangement Question' or SAQ, as coined by Tallant.

Advanced paper: The Argument from Vagueness – Daniel Z. Korman – A modern re-fleshing of Lewis' argument – which I think is one of the most interesting – for mereological nihilism from vagueness.

Advanced paper: The Vagueness Argument for Mereological Universalism – Donald Smith – A discussion of mereological universalism, and in particular, how vagueness challenges and shapes how we understand composition.