Why Is NYT Connections So Difficult?
If you’ve played the New York Times’ Connections puzzle, then you know it is not easy. Not only does it seem like they are trying to trick us (I will never forget, “sponge, “bob,” “square,” “pants”), but the categories are often very obscure. While it’s true that pop cultural and historical knowledge will take a long way, there is another reason (a linguistic one) as to why these puzzles are so hard. It’s something that I haven’t seen discussed much online in conversations about the daily puzzle. It has to do with a very peculiar feature of English, an infinitely productive process called zero derivation. But before we get to that, let’s review a concept you likely haven’t thought about in depth since the 3rd grade: parts of speech.
You probably already know that words can and often do come from other words. As languages develop, grow, and evolve, they take on new features and these combine with those already existent. Across all languages, words exist in groups that linguists refer to as word classes (a fancy term for the parts of speech). Using language requires much more than memorizing words and their meanings; words need to be used in specific contexts which build upon their meanings and situate them in context with others. The parts of speech make this possible. A sentence is the combination of a subject and a predicate, the most common sentence being an actor performing an action.
(1) I walk the dog.
In this sentence, the subject is the first-person pronoun (I) and the predicate is the action performed by that subject (walking the dog). If we wanted to break down the predicate, we can split it into a verb (walk) and an object (the dog). As users of language, our brains understand this without needing explanation. We can instinctually recognize a subject and a predicate even when the meanings of the words themselves aren’t known to us. This is the reason why you can guess with relative certainty what the verb in the following sentences is, even though it doesn’t actually have a definition:
(2) Futvarken dunloked the gerbagertens.
None of those words have a meaning. But your knowledge of English has trained you to pick out the parts of speech. You don’t know who or what Futvarken is, but you know that they dunloked the gerbagertans. The first time you heard someone say, “Can I venmo you?” you might’ve needed to ask what venmo means, but you almost definitely knew it was the verb in the sentence.
So how do we do this? One of the reasons is the order the words come in. While this could possibly be the most reliable way to determine a word’s part of speech, it is wholly unhelpful for a game like Connections where each word receives its own discrete box. One way that we can determine the class a word belongs to while it’s in isolation is to look at its internal structure.
New word alert: affix. An affix is a sound or group of sounds which attaches to existing words. Often, this is in order to (you guessed it) change a word’s class. When an affix joins to an existent word, this is called derivation. The most common affixes are prefixes and suffixes.
Even though you probably couldn’t name them all if you tried, you have hundreds of affixes in your brain that you can use with ease when the occasion arises. These range from the plural suffix /-s/ (cat vs cats), to prefixes like /un-/ (do vs undo), to complex circumfixes (a prefix combined with a suffix) like /em-/ and /-en/ (bold vs embolden). A derivational affix changes one word into a new one with an altered meaning, often of a different word class.
The suffix /-able/ is a common one that changes a word from a verb to an adjective, meaning “able to be [the original verb],” as in ‘compare’ vs ‘comparable.’ The same affix can even change depending on the word it attaches to. For example, a negation affix /-in/ as in “ineligible” becomes /-im/ when it comes before a sound that uses the lips, such as “impossible.” Affixes are built into the language and known by fluent speakers in order to provide them with a way to create new words. This is a process called productivity and it is inherent to all languages.
When verbs become nouns, it is typically rather obvious. The /-er/ suffix is a common one added to a verb to indicate a person who performs that verb. A teacher is one who teaches. A singer is one who sings. If you see the word “singer,” you read it as a noun because it has become one through the added /-er/. Words like this are simple. Unfortunately, most of the English language is not. More on that later.
If you know any other languages, you might be able to think of a typical structure of what a verb looks like, particularly in its infinitive or dictionary form. For example, all three languages I have studied extensively have recognizable verbal suffixes which allow them to be picked out as verbs even when isolated from their contexts.
In Spanish, verbs have three endings: /-ir/, /-er/, /-ar/, or with the reflexive suffix /se/. In Japanese, verbs end in a /-u/. In Russian, verbs end in an /-ti/, /-tʃj/, /-tj/, or with the reflexive suffix /tsja/. If you think about it, it becomes clear that English doesn’t have a clear suffix to indicate which words are verbs. Impressively, almost every sound found in English can be found at the end of a verb.
At this point, those of you who know about the concept of “the infinitive” might have a question. What about the “to,” as in to go, to swim, to eat, to create? Isn’t that functionally a verbal prefix, an indicator that a word is a verb? And to that, I say, it is! Great sleuthing! But the tricky thing is that it is not always necessary and sometimes its inclusion actually makes a sentence ungrammatical. In the sentence, “I ask him to eat dinner with me every week,” the “to” is necessary. In the sentence “I let him eat dinner with me every week,” there is no “to” to be found. In both sentences, “eat” is in its infinitive form.
There is a reason why we don’t find every verb in English under the letter T in the dictionary. That “to” is not part of a verb’s essential form; it arises when it’s needed but it is not there to begin with. One big piece of evidence for this is that an adverb can come between “to” and its verb, as in the sentence, “she seems to really like it.”
When a verb has a recognizable structure like those in Spanish or Japanese or Russian, it suggests that making a verb from a noun or adjective would necessitate a change that matches the verbal structure. And indeed, this is what happens.
In Russian, the word for “vacuum cleaner” is пылесос (pylesos). When one of the verbal suffixes -ить (it’) is tacked to the end, the noun become a verb, пылесосить (pylesosit’) “to vacuum.”
In Japanese, the verb suru する meaning “to do” is very commonly added to the end of a noun in order to turn it into a verbal counterpart. When the word for “studies” benkyou 勉強 takes on the verbal suffix suru する, it becomes the verb benkyousuru 勉強する meaning “to study.”
In Spanish, verbing nouns has become increasingly common in the age of the internet, evident in the creation of new verbs like googlear “to google” and cliquear “to click.”
Now take a look at the following list of 42 English words. When I composed it, I made half of them nouns and half of them verbs. Try to determine which are which (but don’t try too hard).
ship flavor plate wish fish sleep frame dream sign email field scarf name graph race flip track harvest face table spoon corral photograph snowball farm gamble spring chair cramp goal scar bus film purse grade prime foot reward rest guard drink smell
Hopefully you didn’t spend more than a minute looking through that list before you realized that I tasked you with something impossible. Every one of those words can be a noun or verb. Forgive me, I thought it would be a good way to illustrate that English makes it incredibly difficult to tell what class an isolated word belongs to. This is because English, unlike the overwhelming majority of the world’s languages, does not require any affixes to be added to a word when it changes into a verb in a process called zero-derivation. Since verbs don’t have to look a certain way in English, a word doesn’t need to change if it wants to be a verb. It can simply become one, switching spots and changing its grammatical function without a trace.
This gives English a unique ability: almost every noun can be used as a verb without a change in form. This means that the dictionary forms of many verbs are the same as their noun counterparts. Something important to note is that this only applies when the noun is a noun in its most basic form; a noun that was derived from a noun and has a suffix to indicate it will rarely be converted to a verb through zero-derivation when it can just revert back to the verb it used to be. For example, ‘classification’ will likely never be used as a verb since it is a noun which comes from the verb ‘classify.’ Interestingly, though, it still can happen if the noun has gained enough separate meaning. The noun ‘supplement’ comes from the verb ‘supply’ yet ‘supplement’ is also used as a verb with a slightly altered meaning from ‘supply.’ And of course ‘supply’ can also be a noun in addition to a verb, so this is a case of zero-derivation going both ways. As you can see, there are patterns, but patterns don’t always get followed.
Zero-derivation is one of the biggest reasons why English is well-equipped for challenging word-games, the New York Times’ Connections puzzle included. When we see an isolated word like one in the list above, we often will read it and have a gut feeling about if it is a noun or verb. Maybe we use it as a verb more than a noun or vice versa and that meaning sticks out to us more. It’s usually not until we give it a second thought or see it in context with other words that our original notion is challenged.
So all that being said, you made it through the big-brain stuff. Now, it’s time for some linguistic-based tips. You’ve earned it!
How to Hack the New York Times’ Connections Puzzle (according to a Linguist)
Don’t assume it’s a noun or a verb. NYT loves to make use of English’s abundance in zero-derivation. A helpful trick is to put each word in a sentence where it acts as a verb and then as a noun and consider the meanings of both when comparing it to the other words. (Sidenote: if it is a noun that has been derived from a verb, it probably won’t undergo zero-derivation. So you can *probably* skip out on trying to verbify words like government)
Play with the stress. A good amount of seemingly identical verb-noun pairs in English look the same when you see them written, but they are pronounced a bit differently. Technically this is not zero-derivation, instead it is a process called initial-stress-derivation where the stress pattern gets moved to the front of the verb when it is used as a noun. There are a whole slew of words like this, some of the most common examples include: increase, record, convict, rebel, reject, consult, discount, survey, etc.
Is it an adjective? Sometimes, adjectives can become verbs or nouns through zero-derivation. Novel is perhaps primarily thought of as a noun, though its original meaning was as the adjective meaning “new.” Additionally, there are always homonyms. The participle (verbal-adjective) left in “The bike was left on the street” has a very different meaning from the adjective left in “The bike turned left.”
Spelling is not always your friend. On occasion, the categories are chosen based on homophones meaning that the word sounds the same but is spelled differently. If the word has a lot of silent letters, make sure you consider the actual sounds in the word when it is pronounced.
Sometimes spelling is your only friend. Just as a word’s spelling it can screw you over, it can also be your primary clue. I remember there was once a category somewhere along the lines of “Begins with an element’s abbreviation” where the answers were something like “(Fe)line, (Na)ture, (Ag)ain, (Cl)ean.” It could be the case that they want you to take a really deep look at the spelling.
Compounds can do the trick. One category we have seen time and time again is what I’ll call a word-blank category. This is when an unknown word can be compounded with the four in a category. Something like butter____ where the answers are “milk, fly, cup, nut.” These are difficult because often the unknown word doesn’t really relate to the full word; what does a butterfly have to do with butter? But they should always be considered. When you’re stuck, try to see if you can think of any compounds that utilize the words you have left.
Breaking a word into parts might serve you well. When there are compound words given to you, it might be a good idea to break them up and consider their meanings outside of each other. If the word is popcorn, consider pop and corn on their own in addition to the puffy snack.
Slang from all over the world could creep in. There have been several cases of slang from other English speaking parts of the world being included in the puzzles. It might be a game made in NY, but that doesn’t stop the gamemakers from using words outside their dialects!
Proper names are a possibility. It might be the case that the category includes proper names like brands, cities, months, even personal names. A word doesn’t need to have a definition in the dictionary in order to be a contender.
Shuffle. Then shuffle again. This is not a linguistic tip as much as a psychological one, but the shuffle button can be key to giving you the perspective you need to solve the puzzle. I don’t even tend to look at the original grid layout because there is always a trick in there and they try to make it as misleading as possible. The shuffle button is not meant to be a luck tactic; you shouldn’t expect to come upon a row that has all four words from a category. But, you’ll likely find a couple pairs of words you think go together. And always be wary of those pesky false categories!
Until next time :) Thanks for reading!
-Dillon
