Does Thought Precede Language — or the Other Way Around? And So What?
Does Thought Precede Language?
Do ideas and concepts first take shape in our minds, only to be formulated afterward in words and sentences? Or is it the structure of language that directs our thinking? This question has occupied philosophers and linguists for centuries, and continues to do so — even if advances in neuroscience have significantly shifted the terms of the debate. We will see why this relationship between language and thought, and the question of primacy within this pair, carries very concrete consequences for apparently ordinary practices such as dialogue, deliberation, or what we might call everyday political philosophy.
The first position — the idea that we hold mental representations (concepts, images, logical structures) that exist independently of words, and that language is merely a vehicle for expressing them — is generally the most intuitive. But its alternative has its defenders, and its implications are considerably more unsettling : if the structure of language determines, or at least deeply orients, what can be thought, then a language and its specificities have the capacity to influence what individuals think — and by extension, the society around them.
Most contemporary researchers in cognitive science, linguistics and philosophy of mind now avoid both extremes. It is generally accepted that certain forms of thought are pre-linguistic — spatial perception, facial recognition, primary emotions, animal reasoning — which demonstrates that thought is not entirely dependent on language. But for abstract thought — moral, political, metaphysical concepts — language does not merely transport thought : it structures it, and sometimes makes it possible.
The working consensus of contemporary cognitive science might be formulated as follows : thought and language are two distinct systems, but deeply co-evolutionary and mutually constraining ones. For abstract and social thought, language does not carry pre-formed ideas — it shapes them, stabilises them, and defines the contours of what can be thought at all.
And So What? What This Changes
1. What we can think depends on the words we possess — and on their vitality
If language structures abstract thought, then the impoverishment of a society's political and moral vocabulary is not a cultural detail — it is a collective cognitive amputation. A community that loses the word deliberation does not merely lose a technical term. It loses the capacity to think what that word designated : a collective process of forming judgement, distinct from a simple vote or consultation.
This phenomenon resonates directly with what this blog refers to as epistemic contraction — that progressive degradation of our collective capacity to form informed, nuanced and shared judgements. Epistemic contraction does not only strike our relationship to facts and knowledge : it strikes language itself. The polarisation of political debate reduces the available vocabulary to binary oppositions — for or against, with us or against us. Social acceleration eliminates the time needed for words to be weighed, discussed, refined. And this narrowing of language in turn accelerates epistemic erosion — the two phenomena feeding one another in a loop that is difficult to escape.
This is precisely what the philologist Victor Klemperer documented from the inside under Nazism : the regime had not merely used language to propagate its ideas — it had transformed language in order to narrow what could be thought. And the opponents of the regime themselves ended up thinking within their oppressors' categories. The context is radically different, of course. But the mechanism — this capacity of an impoverished language to constrain thought without anyone noticing — remains troublingly relevant today.
2. Discursive framing is a question of power
If words orient thought, then whoever controls the dominant vocabulary controls, at least partially, what a society can conceive of as possible or impossible. Calling a social measure a burden rather than an investment, speaking of migratory flows rather than people in exile, describing a strike as holding people hostage — these are not neutral choices. They are framings that orient judgement before reasoning has even begun.
The linguist Lera Boroditsky has demonstrated experimentally that the way a problem is formulated — even with identical content — modifies the conclusions reached by those asked about it. Which means that in a democratic debate, the question who frames the problem and with what words is just as important as the question who votes.
3. Democratic deliberation as a demanding linguistic practice
This is perhaps the most directly political implication. If collective thought is co-constructed by language, then authentic deliberation cannot be content merely to bring opinions into contact — it must also work on the words in which those opinions are formed.
This implies several concrete things : attention to shared vocabulary — ensuring that participants give the same meaning to key words before debating ; vigilance against imposed framings — identifying when a term already presupposes a conclusion ; and an openness to renaming things — recognising that a problem thought differently may reveal solutions that were invisible within the original framing.
This is not an abstract demand. It is the minimum condition for dialogue not to become a dialogue of the deaf — not because people fail to listen to one another, but because they do not, in the deep sense, speak the same language.
To Conclude — Provisionally
The question of primacy between thought and language is not an academic debate. It touches something very concrete : the quality of our collective life, our capacity to think common problems together, and the resistance we can mount against those who have an interest in keeping our thinking within narrow bounds.
In a context of accelerating epistemic contraction, taking care of language again — questioning the words imposed on us, rehabilitating others we have allowed to empty out, forging new ones when reality demands it — is not an intellectual luxury.
The idea is human. The writing is shared. The exact proportion remains deliberately unspecified.
These texts are published under a Creative Commons license. Feel free to reuse them for non-commercial purposes, and please remember to cite your sources.
