The Spanish Cognates List Is Dead. Long Live the Story.

Cognate lists tell you how many Spanish words you already know. This story puts 854 of them to work. Congo River Journey · Viaje por el río Congo — the densest English–Spanish story we know of, measured.

Here is what that looks like. Three sentences from three chapters, Spanish first, the shared words in bold on both sides. Read the Spanish line before the English one.

Elena filmó el espectáculo desde una distancia respetuosa, consciente de que documentaba no solo una población animal sino un fragmento frágil e irremplazable de nuestra identidad colectiva.

Elena filmed the spectacle from a respectful distance, conscious that she was documenting not just an animal population but a fragile, irreplaceable fragment of our collective identity.

The Bonobo Forest · 100% of content words shared

Cada detalle era crucial: la topografía, la distribución de rápidos, las zonas de vegetación densa.

Every detail was crucial: the topography, the distribution of rapids, the zones of dense vegetation.

The Call of the River · 100% of content words shared

La alteración de sonidos fue total: motores, música, vendedores, teléfonos — un corpus de sonido urbano que vibraba en sus huesos.

The alteration of sounds was total: motors, music, vendors, telephones — a corpus of urban sound that vibrated in her bones.

The City on the River · 92% of content words shared (all but huesos/bones)

Why a story beats a list

Every Spanish course has the list: información, posible, nación. There are about 20,000 English–Spanish cognates to fill it (Nash 1997), and the list proves you know a great many Spanish words. It does not let you read Spanish. Reading means recognizing words in a row, fast enough to hold the sentence in your head while it is still going. The first sentence above is thirteen shared words in a row, and this book has 74 runs of five or more. A list can’t give you that. A story written for it can.

Researchers have suggested the move behind it: to make a text easier for a learner, swap a hard non-cognate for a cognate synonym that looks like the reader’s own word (Hout et al. 2023). We made that swap sentence by sentence, for twelve chapters, in both directions.

The claim, stated carefully

Every chapter of this book exists in English and in Spanish, aligned sentence by sentence. The vocabulary was chosen so that a reader of either language can decode the other without a dictionary.

How much? We counted. First, the ruler.

A “shared word” here means one a reader can see: the same or near-same spelling after the regular correspondences (-tion/-ción, -ty/-dad), and the same meaning in that sentence, measured with the EzGlot similarity metric. Etymology is not consulted; what matters is what the eye can do, and the eye is where Spanish–English cognates are strongest, since they overlap more in spelling than in sound (Dang et al. 2025). False friends count as zero. Names and places are excluded, because Elena and Congo would flatter any text. The method errs low: it misses true cognates whose spelling has drifted (pack/empacar) rather than admit look-alikes.

Chapter one alone measures 73%. Density falls on purpose, from 73% in the first chapter to around half in the last: densest when the reader needs it most, thinnest when they need it least. It doesn’t fall in a straight line, because the story has its own demands (chapter nine, in the city, runs denser than chapter eight before it), but the direction is down.

What the same ruler says about other texts

We measured two public-domain bilingual texts the same way:

Two caveats, because the comparison is only honest with them. A legal declaration is already a Latinate register, so it is the favorable baseline, and the story still measures 63% to its 38%. Cervantes in Ormsby’s Victorian English is the hostile baseline: archaic on both sides. A modern bilingual novel should land somewhere between the two. We would like to measure one; publishers do not release their parallel editions as data.

So the claim is not “the densest ever.” It is: the densest English–Spanish narrative we have been able to measure, by the method described above, and we are asking to be shown a denser one. If you know of one, tell us, and we will run it through the same ruler.

Why density is the number that matters

Comprehension tracks coverage. One of the most cited results in vocabulary research is that readers need to know about 98% of the running words in a text to read it comfortably without help (Hu & Nation 2000), and about 95% for reasonable comprehension (Laufer 1989). Later work found no cliff at either number: comprehension rises steadily with every point of coverage (Schmitt, Jiang & Grabe 2011). Either way, a beginner reading ordinary Spanish fiction sits far below the line, and the fewer words you know, the less there is around the unknown ones to guess from.

Cognates change the arithmetic. A word you recognize from your own language counts as a known word. Learn the function words (a short list, learnable in a week), and roughly four of every five running words in this book are already yours, more in chapter one. That is not the 98% of unassisted reading, and we don’t claim it is. This is assisted reading by design: the other language sits one line below. Density decides how rarely you need it, and the words you do look down for are the ones worth learning.

The recognition is real, and fast. Cognates are recognized and named faster than non-cognates in lexical decision and picture-naming tasks (Costa, Caramazza & Sebastián-Gallés 2000; Dijkstra et al. 1999), and the advantage holds when they are read inside sentences, not just as single words (Van Assche et al. 2009). They are also easier to learn and harder to forget (de Groot & Keijzer 2000). This is the cognate facilitation effect, one of the most replicated findings in bilingual psycholinguistics.

English and Spanish are unusually well placed for it. Those 20,000 pairs are not spread evenly. Lubliner and Hiebert (2011) found that more than 400 of the 570 word families on the Academic Word List have Spanish cognates, about three quarters, which is why the shared layer is the vocabulary of ideas, institutions, science and description that a narrative like this one lives in. Running language is another matter: half the words on the core list of academic spoken English are cognates, yet they cover only about one running word in ten, because the most frequent words mostly are not (Dang et al. 2025). Writing a story that holds 60–70% is the engineering.

For Spanish speakers, this layer is home ground. Of the Academic Word List cognates Lubliner and Hiebert could check for frequency, about half are everyday words in Spanish but academic ones in English: ocurrir and occur, construir and construct, obtener and obtain. Read this book’s English side and you are reading the register that school and work ask for, built from words you already use.

Not every cognate is transparent. Recognition drops the earlier a pair’s spelling diverges. Across the Academic Word List, the later the first differing letter, the more similar bilingual raters judged the pair (r = 0.59) and the faster readers recognized it as a cognate (Hout et al. 2023). Our measurement counts only pairs that survive that test, which is why pack/empacar is not counted even though it is a true cognate, and why the figure errs low.

And the dangerous ones are met, not avoided. False friends are rare, a tiny fraction of core academic vocabulary (Dang et al. 2025), but they are memorable when they bite. Twenty-three of them (sensible, actualmente, compromiso, éxito, constipado, advertencia…) appear in the book, each with a two-sided note: what the word looks like to the English reader, what it looks like to the Spanish reader, what it actually means, and what the misreading costs. A list can warn you that éxito is not an exit. A story lets you walk through the wrong door once, in a scene you will remember.

What the book is

Elena follows the Congo from a modest stream on the Zambian plateau to the Atlantic: river markets, pirogues, manatees, medicinal herbs, two capitals facing each other across the water, the Livingstone Falls. Twelve chapters of adventure, told twice, edited by an award-winning literary translator. Illustrated. Audio in both directions, since words that look alike often sound different, plus a drill track built on graduated recall.

Read the Spanish. When it stops making sense, drop your eye to the English. Do not translate — recognize.

¿Hablas español? Haz lo mismo al revés: lee el inglés y baja la vista al español solo cuando lo necesites.

Chapters 1 and 2 are free, with audio: ezglot.com/books/congo

About Cognata

Cognata is the book imprint of EzGlot, a cognate database spanning nearly a hundred languages, online since 2014 and cited in peer-reviewed research on lexical similarity (Kovačević et al. 2021). The database selects each book’s shared vocabulary; the story is where you find out how much Spanish you already knew.

Deliberately story-sized. A story you finish teaches more than a novel you abandon.

References

• Costa, A., Caramazza, A., & Sebastián-Gallés, N. (2000). The cognate facilitation effect: Implications for models of lexical access. Journal of Experimental Psychology: Learning, Memory, and Cognition, 26(5), 1283–1296.
• Dang, T. N. Y., Macis, M., & Aguilera-Munizaga, M. (2025). Spanish–English cognates and false cognates in academic spoken vocabulary: Lexical coverage, orthographic, and phonological transparency, and frequency. Applied Linguistics. https://doi.org/10.1093/applin/amaf041
• de Groot, A. M. B., & Keijzer, R. (2000). What is hard to learn is easy to forget: The roles of word concreteness, cognate status, and word frequency in foreign-language vocabulary learning and forgetting. Language Learning, 50(1), 1–56.
• Dijkstra, T., Grainger, J., & van Heuven, W. J. B. (1999). Recognition of cognates and interlingual homographs: The neglected role of phonology. Journal of Memory and Language, 41, 496–518.
• Hout, M. C., Montelongo, J., White, B. L., Hernandez, A., & Serrano-Wall, F. (2023). Orthographic similarity ratings for English-Spanish cognates from the academic word list. Frontiers in Education, 8, 1225169. https://doi.org/10.3389/feduc.2023.1225169
• Hu, M., & Nation, P. (2000). Unknown vocabulary density and reading comprehension. Reading in a Foreign Language, 13(1), 403–430.
• Kovačević, L., Bradić, V., de Melo, G., & Zdravković, S. EzGlot. https://www.ezglot.com/most-similar-languages
• Laufer, B. (1989). What percentage of text-lexis is essential for comprehension? In C. Laurén & M. Nordman (Eds.), Special language: From humans thinking to thinking machines (pp. 316–323). Multilingual Matters.
• Lubliner, S., & Hiebert, E. H. (2011). An analysis of English–Spanish cognates as a source of general academic language. Bilingual Research Journal, 34(1), 76–93. https://textproject.org/?p=530
• Nash, R. (1997). NTC’s Dictionary of Spanish Cognates Thematically Organized. NTC Publishing Group.
• Schmitt, N., Jiang, X., & Grabe, W. (2011). The percentage of words known in a text and reading comprehension. The Modern Language Journal, 95(1), 26–43.
• Van Assche, E., Duyck, W., Hartsuiker, R. J., & Diependaele, K. (2009). Does bilingualism change native-language reading? Cognate effects in a sentence context. Psychological Science, 20(8), 923–927.