NEWS.22: `allows' without an object

NEWS.22: `allows' without an object

[Robert J. Chassell]

In English, as heard by a native,

    *** New user option `help-at-pt-display-when-idle' allows to

in emacs/etc/NEWS.22 sounds wrong.  It would sound better if it were

    *** New user option `help-at-pt-display-when-idle' allows you to
                                                              ^^^
Also in that same file, in
    *** You can now use Auto Revert mode to `tail' a file.
the line
    the end of the buffer in that window.  This allows to tail a file:
could be
    the end of the buffer in that window.  This allows you to tail a file:
                                                       ^^^
In
    *** Support for `movemail' from GNU mailutils was added to Rmail.
the line
    This version of `movemail' allows to read mail from a wide range of
could be
    This version of `movemail' allows you to read mail from a wide range of
                                      ^^^

(I think I found all the infelicities involving `allow'.  There may be
others but I stopped because the file was too big.  Mostly, I just
looked at the outline, but even that was large, over 5000 lines.  You
may substitute `GNU Emacs' for `you'.)

--
    Robert J. Chassell                          GnuPG Key ID: 004B4AC8
    bob@rattlesnake.com                         bob@gnu.org
    http://www.rattlesnake.com                  http://www.teak.cc

NEWS.22: `allows' without an object

[Ken Manheimer]

On 5/28/07, Robert J. Chassell <bob@rattlesnake.com> wrote:

> In English, as heard by a native,
>
>     *** New user option `help-at-pt-display-when-idle' allows to
>
> in emacs/etc/NEWS.22 sounds wrong.  It would sound better if it were
>
>     *** New user option `help-at-pt-display-when-idle' allows you to
>                                                               ^^^

well, heck, why not use "enables" instead of "allows"?

"to allow" is to not refuse something, to permit it.  "to enable" is
to activate or provide the means to do something.  seems like 80% of
the time that i see "allows" in instructions, "enables" is what was meant.
(the other 20% of the time, the instructions are properly using the
word to endorse valid inputs, or something like that...)

> looked at the outline, but even that was large, over 5000 lines.  You
> may substitute `GNU Emacs' for `you'.)

"may substitute" - again, permission when something else is meant, in
this case, propriety - "you might substitute `GNU Emacs'..."

i don't mean to claim particular grammar chops, this is just a peeve of mine.
--
ken
http://myriadicity.net

Re: NEWS.22: `allows' without an object

[Eli Zaretskii]

> Date: Mon, 28 May 2007 17:17:10 -0400 (EDT)
> From: "Robert J. Chassell" <bob@rattlesnake.com>
> 
> the line
>     This version of `movemail' allows to read mail from a wide range of
> could be
>     This version of `movemail' allows you to read mail from a wide range of
>                                       ^^^

I think "allows reading mail" is also okay, and doesn't require "you".

Re: NEWS.22: `allows' without an object

[Richard Stallman]

Thanks.  There must be someone who writes entries
for etc/NEWS who thinks that is a valid English construct.
If he reads your message, he will probably know better in the future.

Re: NEWS.22: `allows' without an object

[Alan Mackenzie]

Hi, Robert!

On Mon, May 28, 2007 at 05:17:10PM -0400, Robert J. Chassell wrote:
> In English, as heard by a native,

>     *** New user option `help-at-pt-display-when-idle' allows to

> in emacs/etc/NEWS.22 sounds wrong.  It would sound better if it were

>     *** New user option `help-at-pt-display-when-idle' allows you to
>                                                               ^^^

This is a very common solecism in all technical English written by
"foreigners";  though I thoroughly respect, even admire, their mastery of
such a bastard capricious language as English.

I think it would be useful to emphasise the principle: "allow" needs a
DIRECT OBJECT.  This d.o. can either be:

(i) the person or thing being empowered:

  "auto-revert allows YOU to 'tail' a file";
  "auto-revert allows EMACS to tail a file";
  
(ii) The object of the allowed action:

  "auto-revert allows A FILE to be tailed".
  (this is really the same as (i), but with a passive verb);
  
(iii) the process being allowed - this is often a gerund[*], but need
not be:

  "auto-revert allows THE TAILING of a file";
  "The law courts allow THE FILING of a tale";

[*] "gerund", a grammatical term, means the noun form of a verb: "the
allowing".  It has the same form as the present participle, but is
distinct from it.  Other languages just use an infinitive here - e.g. in
German, "das Erlauben".

The same (or very similar) constructs are used by many other English
words: "enable", "permit", "encourage", "force", "help", "suggest", ....

Also to be noted: negative words like this ("prevent", "discourage",
...) tend to use "from" rather than "to":

  "Write protection prevents you FROM altering a file."
  "Write protection prevents a file FROM being altered."
  "Write protection prevents the alteration of a file."

Like all grammatical rules in English, this one has exceptions.  We
English speakers have got to keep presumptious foreigners in their place
somehow.  ;-)

-- 
Alan Mackenzie (Ittersbach, Germany).

Re: NEWS.22: `allows' without an object

[Stefan Monnier]

>> the line
>> This version of `movemail' allows to read mail from a wide range of
>> could be
>> This version of `movemail' allows you to read mail from a wide range of
>> ^^^

> I think "allows reading mail" is also okay, and doesn't require "you".

That's indeed what I was told, and that's a source of the confusion (for me
at least): in most contexts, "blabla to read mail" and "blabla reading mail"
translate into the exact same thing in French, so I tend to not know when to
use which.
In the specific case of "allow" and "enable", I know I'm not alone ;-)


        Stefan

Re: NEWS.22: `allows' without an object

[David Reitter]

On 29 May 2007, at 09:48, Stefan Monnier wrote:
>> I think "allows reading mail" is also okay, and doesn't require  
>> "you".
>
> That's indeed what I was told, and that's a source of the confusion  
> (for me
> at least): in most contexts, "blabla to read mail" and "blabla  
> reading mail"
> translate into the exact same thing in French, so I tend to not  
> know when to
> use which.
> In the specific case of "allow" and "enable", I know I'm not alone ;-)

"Reading" is used as a noun here, and "allow + noun" is an English  
construction. Additionally, English allows for the omission of the  
agent of nouns that have been converted from verbs (read -> reading):  
"Reading was popular in the 50's" is okay, even though it is not  
specified who is reading. (The genitive takes the place of the  
subject, otherwise: "Peter's writing was awful.")

"Allow" is a control verb, where the entity that experiences the  
"allowing" is realized as an object, but is also the subject of the  
complement clause: "A allows B to do X" means "A permits that B do X".

A complement with "to" always means that a clause follows, and this  
calls for a subject in English. With a control verb like "allow", you  
should therefore always use the extra object: "The rain allows John  
to go back to bed."

In the general case, I find the verbal style better than the noun  
style, but native speakers may have a different view.



--
David Reitter
ICCS/HCRC, Informatics, University of Edinburgh
http://www.david-reitter.com

Re: NEWS.22: `allows' without an object

[Robert J. Chassell]

As Alan Mackenzie <acm@muc.de> says, "allow" needs a direct object, 

   >     This version of `movemail' allows you to read mail from a wide range of
   >                                       ^^^

   I think "allows reading mail" is also okay, and doesn't require "you".

`Reading' serves (or maybe the object is the whole phrase, `reading
mail' -- I don't know.)

Stefan Monnier <monnier@iro.umontreal.ca> wrote

    ... in most contexts, "blabla to read mail" and "blabla reading
    mail" translate into the exact same thing in French, so I tend to
    not know when to use which.

I did not know.  That explains a great deal.

As Alan Mackenzie says, this instance needs `the person or thing being
empowered'.  On its own, in English, the phrase `to read' fails.

The English is confusing.  It may be that you can only comfortably
learn this kind of construction when very young.

You could write, `enables reading mail', too; that makes more sense.
Before Eli Zaretskii made this observation, I had not noticed the
distinction between gaining permission and gaining an ability, but it
is there and important.  After all, we are not talking about humans
getting permission from the `movemail' code, as `allow' suggests, but
gaining from it the power to act.

-- 
    Robert J. Chassell                          GnuPG Key ID: 004B4AC8
    bob@rattlesnake.com                         bob@gnu.org
    http://www.rattlesnake.com                  http://www.teak.cc

Re: NEWS.22: `allows' without an object

[David Kastrup]

"Robert J. Chassell" <bob@rattlesnake.com> writes:

> As Alan Mackenzie <acm@muc.de> says, "allow" needs a direct object, 
>
>    >     This version of `movemail' allows you to read mail from a wide range of
>    >                                       ^^^
>
>    I think "allows reading mail" is also okay, and doesn't require "you".
>
> `Reading' serves (or maybe the object is the whole phrase, `reading
> mail' -- I don't know.)
>
> I did not know.  That explains a great deal.
>
> As Alan Mackenzie says, this instance needs `the person or thing being
> empowered'.  On its own, in English, the phrase `to read' fails.
>
> The English is confusing.  It may be that you can only comfortably
> learn this kind of construction when very young.
>
> You could write, `enables reading mail', too; that makes more sense.

Actually, I find that "enables" is suffering from a similar degree of
awkwardness.  I'd probably use "facilitates reading mail ..." instead:
this is an enabled-object-free verb, though a bit more pompous.  More
closely related to "allows" would be "permits reading"; this is
simpler than "facilitates", though, like "allows", slightly wrong as
this is not a question of permission.

> Before Eli Zaretskii made this observation, I had not noticed the
> distinction between gaining permission and gaining an ability, but
> it is there and important.  After all, we are not talking about
> humans getting permission from the `movemail' code, as `allow'
> suggests, but gaining from it the power to act.

I should read postings to their end before replying.  Saves time.

-- 
David Kastrup

Re: NEWS.22: `allows' without an object

[Ken Manheimer]

On 5/29/07, David Kastrup <dak@gnu.org> wrote:
> "Robert J. Chassell" <bob@rattlesnake.com> writes:

> > You could write, `enables reading mail', too; that makes more sense.
>
> Actually, I find that "enables" is suffering from a similar degree of
> awkwardness.  I'd probably use "facilitates reading mail ..." instead:
> this is an enabled-object-free verb, though a bit more pompous.  More
> closely related to "allows" would be "permits reading"; this is
> simpler than "facilitates", though, like "allows", slightly wrong as
> this is not a question of permission.

not sure what you mean by "enabled-object-free verb", but i often find
"facilitates" a lot more stuffy than "enables".  in any case, they
both avoid misuse of "allows".

> > Before Eli Zaretskii made this observation, I had not noticed the
> > distinction between gaining permission and gaining an ability, but
> > it is there and important.  After all, we are not talking about

(i think robert is referring to the point i made, not eli?)

> > humans getting permission from the `movemail' code, as `allow'
> > suggests, but gaining from it the power to act.

exactly.  this is one of those subtle misuses which continues because
it is common, yet (i doubt) is part of gradual language evolution,
because relaxing the meaning of "allow" so much would render it
ambiguous.
-- 
ken
http://myriadicity.net

Re: NEWS.22: `allows' without an object

[Eli Zaretskii]

> From: Stefan Monnier <monnier@iro.umontreal.ca>
> Date: Tue, 29 May 2007 04:48:50 -0400
> Cc: bob@rattlesnake.com, emacs-devel@gnu.org
> 
> In the specific case of "allow" and "enable", I know I'm not alone ;-)

"Enable" is almost always wrong in such contexts, so it is better to
avoid that word altogether.