I was reading this article


when I saw

Propylene Glycol is controversial, and is said to may be potentially harmful to your health

As far as I know, "may" cannot work as an infinitive like this, but I decided to try to Google for it, and found other cases, e.g.


Now it is Iran that is said to may be bent on acquiring nuclear arms, and President Bush who has declared that "unacceptable."

I'm beginning to think that these are not mere typos. But what's going on? I tend to believe that the authors are native English speakers, at least I didn't find any other signs of non-native-ness in the articles. Is it dialectal?

Please note: I am not looking for answers along the line of "This is wrong".

  • 5
    It is not formally correct; may, like all modal auxiliaries, is defective, having no non-finite forms and no person/number inflections. But unlike can (be ableta), must (hafta), and will (be gonna), may has no non-defective periphrastic synonym, so it doesn't surprise me to find folks trying to recategorize its forms. I don't think it will catch on, but I won't live long enough to find out. – StoneyB on hiatus May 27 '14 at 12:41
  • 4
    "May" does not have an infinitive. I really do think these are typographical errors. I am not familiar with any dialect in American English in which "to may be bent on acquiring" would be considered standard. It seems obvious that "is said to may" is an editing mishap, in which "is said to" was inserted to replace "may," but the writer forgot to delete "may." – outis nihil May 27 '14 at 12:46
  • @StoneyB, the lack of a non-defective periphrastic synonym is an interesting theory! But can't you usually use "possibly" instead (although "possibly be potentially" seems quite clumsy)? – dainichi May 27 '14 at 12:52
  • 9
    It looks to me like an erroneous contraction of "...said to maybe be bent on..." – Rupe May 27 '14 at 12:54
  • 1
    I had the same thought as @Rupe. The author may have thought that maybe be sounded clumsy, and didn't realize that his correction was just as poor. – Barmar May 30 '14 at 16:30

Occurrences of “to may be” in the form that the OP describes are rare in a Google Books search of books published between 1600 and 2009, but I did find two matches, from books published in 2006 and 2009. From Roopchan Lutchman, Sustainable Asset Management: Linking Assets, People, and Processes for Results (DEStech Publications, 2006):

We have heard the old sayings “What gets measured gets done” and “Without data you are just another opinion” many times and they may seem to may be simplistic but has real meaning in the business environment. The best of strategic plans and intentions can amount to very little unless there are clear goals, objectives and associated targets to ensure that strategies for success are being achieved.

From T. Hayashi and A. Myakoshi, “Land expansion with reclamation and groundwater exploitation in a coastal urban area: A case study from the Tokyo Lowland, Japan,” in From Headwaters to the Ocean: Hydrological Changes and Watershed Management (CRC Press, 2009):

Confined groundwater in the reclaimed areas shows different chemical properties with the inland part of the lowland. Chloride ion, hardness and potassium permanganate consumption value of groundwater in the reclaimed area are higher than those in the inland part. Considering the hydrological setting and the distribution of these components, there is no good reason to think that these components had been recharged to groundwater from the ground surface of the reclaimed area. Groundwater in the reclaimed area is considered to may be originally in the stagnant condition. The result of this study suggests that it is possible to use groundwater beneath the coastal seafloor as water resources, but more careful and sufficient evaluation of groundwater environment is required because groundwater in this area may be in stagnant condition.

Neither of these instances of “to may be” seems unintentional and neither involves a simple error of splitting “maybe” into two words; but the meaning in both instances would have been clearer if the authors had used the wording “to maybe be” or had reworked the relevant sentence to express the authors’ uncertainty more elegantly.

On the other hand, the phrase “to maybe be” is itself a fairly recent arrival in published writing. Easily the earliest match in a Google Books search is this instance from Anne Warner, Susan Clegg and a Man in the House (1907):

”Not me,” said Miss Clegg; “I ain’t got any give-up in me. I’ll keep on until I find it if I have to board Elijah Doxey till he dies or till I drop dead in my huntin’ tracks. But I see that my feelin’ towards him is n’t goin’ to be what it might have been if he’d been frank an’ open with me as I am with him an’ every one else. He seems frank an’ open, too—in other ways than that box. He read his editorial aloud night afore last an’ I must say it showed a real good disposition for he even wished the president well although he said as he knowed he was sometimes goin' to be obliged to maybe be a little bit hard on him. …”

The next matches for “to maybe be,” stretching from 1949 through much of the 1990s, occur in transcriptions of recorded testimony or in fictional dialogue. In fact, the earliest occurrence that a Google Books search found of “to maybe be” in a scholarly setting was from Colette Grinevald, “Living in Three Languages,” in Essays on Language Function and Language Type (1997):

One of my first courses in linguistics that Fall was phonetics, with Steven Anderson. I was grateful to be introduced to phonetic transcription early on, as I did not understand too much of the lectures. I took notes in French and started writing down phonetically those words that seem to recur often enough to maybe be of some importance, linguistically or perhaps just in general. I read those phonetic transcriptions to my household at dinner time.

On this evidence, it appears that “to may be” remains a very rare alternative formulation for the phrase “to maybe be,” which itself is far more common in informal speech than in formal (and copyedited) writing. For the time being, “to may be” is likely to sound like a mistake to most English speakers; but years from now, authorities may be pointing out that the wording has been attested in serious writing “since at least 2006.”

Related Historical Note

In the course of my Google Books searches, I was surprised to find a specimen of “to may be” in a treatise evidently written during the reign of Henry VI. From John Fortescue (who died circa 1480), The Difference Between an Absolute and Limited Monarchy ; As It More Particularly Regards the English Constitution:

And if it happyn that any Patent be made of any parte thereof [that is, of the King’s ordinary charges] to other use, that than the Patent to be voyde, and of non effecte : Which thyng, yf it be fermely establyschid, the Kyngs Ordinary chargys may alway be paid in hand, and the Provysyon of them may be alway be made in season ; which schal be worth to the Kyng the fourth or fifth part of the quantite of his expenss for his Ordynarye charges. This may in nothyng restrayne the Kyngs Power. For it is no Power to may alien, and put awaye : But it is Power to may have, and kepe to hymself. So it is no Power to may syne, and to do yll, or to may be syke or wex old, or that a Man may hurt hymself. For all thees Powers comyne of Impotencye.

In the 1714 edition of this work, edited by a descendent of the author, John Fortescue Aland, the editor provides the following footnote in connection with “to may alien”:

To may alien, to may have, i. e. to be able to alien, and to be able to retain, from the Saxon Verb, magan, posse, to be able ; which see before, in the word may.

In the earlier note, the editor observes that

mæg is the present Tense of the Saxon Verb magan, which signifies to be able, or to may, do a thing, as old Authors express it.

In this older understanding of "to may be," it seems, the sense was not "to maybe be" but "to be able to be."


Something I find a little strange about this usage is that "may" initially appears redundant:

Propylene Glycol is controversial, and is said to may be potentially harmful to your health

Propylene Glycol is controversial, and is said to be potentially harmful to your health

But technically there is a difference between "could potentially harm you" and "could not harm you". Using "may", here, is stating that propylene glycol may be in the former category. It may be possible that it could potentially harm you.

Noting that it is said that it may be possible that it could potentially harm you is pushing it out one more layer:

[Propylene Glycol is said to] [may be [potentially harmful to your health]]

Someone says that it may be potentially harmful. You could rearrange the sentence as such:

It is said that Propylene Glycol may be potentially harmful to your health

Moving "Propylene Glycol" to take the place of "it" and replacing "that Propylene Glycol" with "to" is what makes the sentence sound strange.

[It] is said [that Propylene Glycol] may be potentially harmful to your health

[Propylene Glycol] is said [to] may be potentially harmful to your health

Let's see if your second example does the same thing:

Now it is Iran that is said to may be bent on acquiring nuclear arms, and President Bush who has declared that "unacceptable."

It is Iran that is said to may be bent on acquiring nuclear arms

It is said that Iran may be bent on acquiring nuclear arms

Almost. This usage keeps "it is" and "that" but we could backstitch that into the "Propylene Glycol" example:

It is Propylene Glycol that is said to may be potentially harmful to your health

In both of these examples, the traditional variant would use "maybe be" instead of "may be":

Propylene Glycol is controversial, and is said to maybe be potentially harmful to your health

It is Iran that is said to maybe be bent on acquiring nuclear arms

But the "be be" sound sounds really awkward and it doesn't surprise me that it was dropped. The writers probably felt that the awkward sound of "may be" was less evil than the awkward sound of "maybe be". Since the alternative forms of both examples use "may" anyway, it doesn't seem that far of a stretch.

Unfortunately, however, I am not an expert on parts of speech. I couldn't tell you what the difference between all of these moving parts is. Hopefully this helps somehow, anyway.


'May' in each case is not the infinitive but a verbal auxiliary. The constructions as a whole are examples of split infinitives which some (including me) still find awkward.

There is no need for an infinitive at all. They could easily have been written:

a) Propylene Glycol...may, it is said, be harmful to your health,and

b) Now it is Iran that, it is said, may be bent on acquiring...'.

However may could easily be replaced with an adverb such as perhaps - which is more flexible so far as avoiding split infinitives is concerned.

  • 1
    AFAIK, "split infinitive" is a term used when adverbs appear between "to" and the infinitive. I've never heard it applied to auxiliaries. – dainichi May 27 '14 at 13:00
  • 1
    This doesn't actually answer the question. How is this a split infinitive? – Peter Shor May 27 '14 at 13:16
  • @dainichi Well 'may' is acting as an adverb in a sense. Try replacing it with 'perhaps' in the OP's examples. – WS2 May 27 '14 at 19:35
  • @PeterShor It splits the 'to' from the 'be'. – WS2 May 27 '14 at 19:41
  • But for most split infinitives, you can unsplit it by moving the adverb to one side or the other: "to boldly go""boldly to go" or "to go boldly". If you try it here, you get "is said may to be potentially harmful to your health" or and "is said to be may potentially harmful to your health", neither of which is any better. – Peter Shor May 27 '14 at 20:33

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy

Not the answer you're looking for? Browse other questions tagged or ask your own question.