I read a lot of books as well as reviews.
I also write both books and reviews.
Reviews and Ratings are the currency by which books are valued in the book industry — but this value is only a reflection of reality if readers take the time to write reviews, or at least give a rating and upvote any reviews that are helpful, fair and capture their own views (rewarding the good reviewers). Of course it important that reviews don't give away things that should come as surprise (fiction) or would give away secrets (non-fiction/how-to).
I try to be guided by the author/publishers' own description and blurb, what they themselves reveal. But some newer Indie authors do give away too much, particular in descriptions of sequels or multibook box sets (so avoid reading beyond the description of the first book). Recently I wrote a review in which the Subject/Title was a warning not to read too much of the author's description, but since the dots for reading more come before the reviews on Amazon, most will read that before getting to the reviews.
Me? If a description seems to giving away the plot, I will simply click on the stars/ratings link and have a look at the reviews - again skipping forward quickly if I sense any spoilers coming. Another approach is to start on a review site like Goodreads and look for the highly rated books in your genres of interest. Many reviewers post on both Goodreads and Amazon (and indeed getting to the last page of an Amazon eBook should trigger opportunity rate/review on both with one click. Professional or author reviews who review on other sites or their only blog should do so too (and you shouldn't have to ask).
As reviewers, there are many things we have to balance, and the balance will vary depending on the purpose of the review, and the type of review.
Editorial Reviews, Advance Reader Copies and Galley Proofs
The first kind of review is the editorial review, written before a book is published, by a 'professional' reviewer. This is intended to supply the kind of endorsements and comments that might appear on the book cover or on its online sales/preorder page, and thus is obtained well before publication using a preliminary version of the book (ARC or Proof).
Newspapers and magazines, blogger and websites are the major sources of such reviews. Professional can mean one of two things, one that the person has written a lot of reviews (experienced), or two that the reviewer has been independently certified in some way (those associated with a formal publication or reputable company).
Advance Reader Copies (ARCs) are often made available by authors on an ad hoc basis (or through sites like BookFunnel). Especially for traditionally published books, they are made available on sites like NetGalley and Edelweiss that are dominated by traditional publishers, make ARCs available to reviewers from the galley proof stage, and vet the reviewers as professional both in the sense of having experience and having an outlet.
Another major source of early reviews is authors of similar books — publishers like to get quotable endorsements from writers in their stable. I am always wary of these, because often the words are put into the mouths of those authors by the publishers, and that author may not have even read the entire book. I won't write an endorsement or positive review unless I have (though publishers tend to make their publishing decisions on a synopsis and an opening chapter).
There are also websites that allow you to buy such review and endorsements. The reputable ones seek to be 'honest'. Some try to show they are honest by being 'balanced' - and this is often be interpreted to mean that they must include some negatives in a review. Unfortunately, these can be forced or even totally made up ("well it is self-published so there must be some typos and grammos even if I didn't notice them").
If it is worth mentioning typos and grammos in a review, then it is important to provide examples. Without evidence, such a statement is worthless even libellous - and for an ARC review there will be a path to give specific feedback to the author or publisher if the reviewer is inclined to be helpful and there aren't too many errors. So it is important to address the actual issues clearly and accurately (I take notes as I read), to be clearly helpful and obscurely diplomatic... What ever you do, don't attack the author or say anything that can be construed as such! Read it from the author's perspective before hitting that commit button.
Unhelpful and Misleading Reviews vs Helpful Reviewers
One reviewer (from a company deliberately not mentioned in this blog post) said of one of my ARCs/Galleys "The story is marred somewhat by formatting mistakes, grammatical errors, misused words, and other linguistic details, but the heart of the story is strong, and the premise is enchanting from the start."
The company would not retract that, but did correct the repeated use of a masculine pronoun for my decidedly female heroine (who was one of a group of middle grade girls, and at one point was asked to take off her bra for a medical scan). The Editor-in-Chief did take the time to scan the book, and pointed out deviations from their formatting expectations (clearly based on the much deprecated Chicago Manual of Style), ignoring the fact that this was a prepublication ARC/galley/proof provided them for prepublication editorial review purposes. However, my query explicitly focused on the "grammatical errors, misused words, and other linguistic details" and these remain unidentified: a new read through the book confirmed that this statement was unwarranted (not to mention all the previous input I've had on the book).
So I am left with a statement, in multiple places on the web, that impugns my reputation as a Professor of Computational Neuroscience and Psycholinguistics writing a story that builds on this specific expertise, uses the appropriate terminology, refers to actual biomedical equipment and scientific theories using the appropriate terms, and sometimes even names the brand of the device concerned. Frankly, I am dubious that the reviewer really read the book (missing that the protagonist was a girl), but maybe they did a quick scan like the EiC and saw some words they didn't understand. Needless to say, I am disinclined to trust my prepublication ARCs/Proofs to this company again for editorial review.
On the other hand I, as a reviewer, have to make sure that I don't fall into the same trap. Especially as you can't argue with anonymous reviewers, and authors really have no recourse. Anyway, as an author/reviewer I want to help authors, particularly indie authors, not make problems for them.
If you are dealing with a prepbulication ARC/galley/proof, errors should be communicated to the author or publisher so they can be fixed. You are not dealing with the final format of the book, so complaints about format are inappropriate and unprofessional. Personally I dislike the double-spaced formats that are expected in manuscripts (or more correctly typescripts) per the traditional and modern formats specified by most publishers and agents, and some review services. Note that manuscripts are, etymologically, handwritten handscripts, and that the wide-margin double-spacing convention in typescripts and thesis submissions was primarily so that proofreaders and examiners could write corrections between the lines, with proofreading marks in the margins. The secondary advantage, for proofreaders, is that it slows you down as a reader, stops you getting too immersed in the story, and focusses you on the minutae of formatting, spellings and grammar; whereas the final format of the book is designed to guide your eye through the text quickly and make reading easier.
As reviewers, we are not paid proofreaders, and are meant to be appreciating the work as whole, getting the same big picture that a normal reader will have. Double-spaced PDFs (or ePubs - yes I got one of those recently) are totally inappropriate.
On the other hand, supplying editorial reviewers in a double-spaced format might avoid the kind of complaints about formatting that I quoted above. But as a reviewer, I avoid these double-spaced copies, and indeed PDFs in general (if permissions allow, I will tend to convert such PDFs or DOCs to single-spaced versions, ideally ePubs, that can be read comfortably on my eReader — if permissions don't allow this, I have on occasions declined to read things too unwieldy to read fluently on my eReader or been blocked from reading them).
In particular, sometimes books are provided with Digital Rights Management as the encrypted ACSM files that can hide different formats, and the app that opens it doesn't necessarily know how to deal with that hidden format, the once-only DRM locking you out for ever. Edeweiss is the big culprit here as it does not tell you what formats are availabe, while NetGalley does and prioritizes ePubs. Also ePubs are sometimes faked by taking page images to make an illegible ePub that doesn't scale to a handheld device; moreover Adobe Digital Editions (ADE) is not suitable for reading on tablets/phones without keyboard/mouse. However NetGalley also has its own app in which selected and approved or autoapproved ARCs magically appear, which navigates with simple left/right swipes like the Kindle app.
So ARCs tend to be ePubs or PDFs in galley proof form, in which case they should not be double-spaced, and may or may not approximate the final publication (galley proofs exist primarily so that final proofreads can get rid of the remaining typographical errors, while ARCs are for advance/beta readers and editorial reviewers — but these days the same document usually serves both roles). They may also be physical books, e.g. Proof or Author copies sent direct from Amazon, or similar copies obtained from other PoD publishers like Direct2Digital.
[Also be aware that (Word 365) DOCX reduces embedded images to about 200dpi, irrespective of settings, which can impact the quality of images/cover and thus reviews. The older DOC format doesn't compress and is much safer (just don't make a DOCX and then use that saved version to create a DOC or PDF, as the loss is by then already irretrievable). I have also stayed with the non-X version for posters in Powerpoint for similar reasons (and keep several older versions of Microsoft Office installed on one of my computers to deal with the various other incompatibilities and downgradings of the last 20 years).]
Children's Books: Picture Books through Middle Grade to Young Adult
I am always happy to send either ePubs or paperbacks to reviewers (and for children's books the reviewer may actually mediate/mentor a child's reviews - yes, I'm happy to send them to parents/teachers who write reviews for their children to look at, and for children's books physical books may be necessary/preferred).
At this point it is worth mentioning the Wishing Shelf Awards and Reviews, out of Europe (UK+Sweden), which actually gets groups of school children to review of appropriate age groups to review Children's/MG/YA books - and provides good feedback if requested. They also have adult reviewers reviewing the full range of books, so don't let the name put you off.
Incidentally, beta readers and beta testers refer to a second phase of testing by people removed from the creative process — the first, or alpha, stage of reading/testing is done in house. Again, I try to include children in a broad age range around my target audience. It is important to incorporate feedback from alpha and beta readers of the appropriate demographic before sending books out for editorial review.
A writers group can be a good source of readers — in fact, for my writers group it is in the first instance alpha hearers, although later those who are interested will have opportunity to feed in at the galley/proof/beta stage. But we are still second guessing what the children will think.
Children live in the adult world and hear and watch adult conversations, adult news broadcasts and adult-oriented TV shows. And just because a child is in a certain class or of a certain age doesn't mean they are the same as every other child of that class/age. So explore, ask children to read your ARCs — and ask them which of their friends would enjoy it!
"Friends, neighbours and countrymen" can be a good source of parents and hence kids — and yes, "lend me your ears": reading the book out loud, or segments of it, can be useful. This is especially the case when your audience is MG/YA or younger.
Teachers, parents and librarians and publishers have a fair idea of what children will like, but these days some gatekeepers tend to play down to a perceived low level expectation or up to a particular agenda, rather than encouraging young readers to explore and grow and develop, raising their reading level, developing their vocabularies, enhancing their cognitive capabilities, and learning social skills.
Publication of Editorial Reviews
Editorial reviews, or their snappy one liners, glowing phrases and explicit endorsements, often find themselves on the cover of the book or in the description on the bookshop websites. Amazon's author page also allows you to put these into a separate section on your Amazon Author Page (which is used on their main North American .com site, but not on others).
The full reviews are generally published in the reviewers' own media outlets, whether website, blog or physical magazine or newspaper.
Many reviewers will also (optionally or automatically) put them up on Goodreads and/or Amazon. Some allow you to opt out, or have it go up only if it gets a certain number of stars. I always allow it to go up, and don't select the 'suppress if bad' option. In fact, every review helps (the number of reviews is important) and a spectrum of reviews reflects different people having different tastes and different expectations. But this is where upvoting is important (or downvoting if allowed) as this affects which reviews are shown first, even if filtered by number of stars. In some cases, commenting is also possible (ideally by readers rather than the author).
Reviewers should acknowledge it if they received an ARC, and specify how (from the author, publisher, Netgalley, Edelweiss, ...), and should then go on to explain that it was voluntary. I tend to say something like: I received an obligation-free copy of this book from the publisher and am reviewing it voluntarily — my opinions are my own.
Amazon Verified Purchaser vs Kindle Unlimited - Customer/Reader Reviews
While reviews of books you receive for free can, and should, be included by the reviewer amongst the normal reviews on the book pages, Amazon makes special note of 'Verified Purchaser' reviews. That is people who actually spent money to get the book. This doesn't include ARCs, and unfortunately it also doesn't include borrows on Kindle Unlimited — even though we have paid for the privilege and authors get paid for each page read (typically close to half a cent a page, which is actually about half what they'd get for a 300 page eBook at the minimum price of USD2.99, although actually more like a third of the per page printing cost for pBooks).
There are websites that encourage authors to review books in a pool, and carefully develop reviewers skills and rate or certify their reviewers.
A long established one is OnlineBookClub, although its website is rather clunky. It is strongly moderated, meaning there can be delays due to the moderation process as they sort out discrepancies (which is good for authors and extra work for reviewers). Authors pay different amounts for the editorial reviews according to the level of the professional 'team' reviewer (who gets paid). Authors can't be team reviewers, but once a book is team-reviewed, the Authors Review Authors (ARA) scheme allows authors to gain credits for reviews of other people's books they've completed and then spend these on getting a review from the pool of ARA members.
CES Pro has a similar, but much newer, Community Book Exchange Program where authors can add a book to the program for review and in turn be given a book to review (they try to make it relevant, and you can request another).
In both cases, you obtain the book in the normal way, and if available for a dollar (e.g. on an Amazon Countdown) I will buy the book rather than borrow it, where it would take up limited space in my KU library and the review would forgo the coveted 'Verified Purchaser' label. For a 600 page book at the minimum price of $2.99 or a 200 page book at the countdown price of $0.99, the author gets around the same through Kindle Unlimited/Kindle Select as for a sale; for a shorter book, proportionally less.
Note that, in both OBC's ARA and CESP's CBEP, these are not book swaps where authors review each other's books. This is regarded as unethical, and generally sanctions will be applied if discovered (e.g. by Amazon). An important principle is that reviewers are at arms length and reviews are voluntary but this doesn't mean you can't know an author you're reviewing (e.g. personally/well, through professional correspondence, or by meeting them at a conference or fan meet).
What about if you are reviewing a book in return/credit for reviewing a third party book from a review pool? This meets Amazon's guidelines: OBC ARA shows you 50 possible books at a time, and allows you to select up to five at a time to look at, and CESP CBEP requests you to review one. But if you don't like the book, and don't feel that you want to waste more time on it — and it is not that it is a bad book, just not to your taste — in both reviewing pools you can opt out and get another book to review.
Feedback and Errata: English Grammar, Spelling and Formatting
Unless a book is really unprofessionally presented, I would normally make no comment in the review about occasional typos, grammos and formatting errors in the book, but if possible will pass them back to the author (through OBC or CESP if they nominated the book). But more generally if reviewing directly on Amazon or Goodreads, this direct author contact is not easy (unless there is contact information for the author in the book). In such cases I might mention the errors and give one or two examples of each kind.
There are, however, some things I find really annoying but have to hold back on, e.g. routine use of "and I" when it is the object of a verb or preposition, and thus accusative, e.g. in constructions like "He gave my brother and me a lift", "He then wanted my brother and me to pay for the petrol", "In the end he got physical and demanded money from my brother and me." Incorrect using "and I" is okayish in dialogue where you are trying to convey that the speaker is uneducated and trying to obey the grammar rules of the upper crust, but best practice is not to get too carried away conveying dialect. The educational problem here is that in response to the argot "me and John went to the park" parents/teachers simply say "No, don't say 'me and John' say 'John and I went to the park'" and themselves don't know and explain the rules (two here: the grammatical nominative/accusative distinction; and the social convention of putting yourself last). This is called overgeneralization — the rule the children learn is always say 'and I' and never 'me and' or 'and me', and then it may be further generalized with 'and him' becoming 'and he', etc.
In relation to formatting and spelling, reviewers need to be aware that there are different models, and while US publishers can insist on Webster and the Chicago Manual of Style, there are other countries and other conventions — and for the US, the New York Times Manual of Style is much much better than the outdated Chicago one that was originally dictated by printers on the basis of what looks good, rather than sound principles of grammar and logic. Of course, rather ironically, modern British spelling, grammar and punctuation was largely standardized by William Caxton's publishing house in the late 1400s, in the name of consistency — presumably some authors railed against that too when their preferred forms were deprecated. I would never complain about such conventions in a review just because it looked strange or unusual to me. But if it really is a reason not to buy the book, if it really makes the story hard to understand, then breaking conventions is worth mentioning — with explict explanations of the specific issues you are concerned about.
One bugbear in relation to punctuation is the em-dash — which is meant to be the width of an 'M' (and in modern typesetting conventions should have space around it, but perhaps narrower spaces than usual) while the en-dash (the width of an 'N' with normal spaces around it) is also accepted/recommended. No spaces means that it combines with the words on either side to look like a single word, and word processors treat that combination as a word. Poorly (or 'artistically') designed fonts can have very long em-dashes; and long dashes, small spaces or no space cause problems for the word processor in trying to break lines typeset (left and right) justified text. Dashes are used as weaker form of parenthesis than brackets — often with explanation or examples — but stronger than commas, and less ambiguous when there are commas elsewhere in the sentence. When the closing bracket would be at the end of the sentence, comma and dash parenthesis is still quite appropriate — it is just that there is no explicit closing punctuation needed as the sentence terminating punctuation terminates it too (as does nesting enclosure by actual parentheses).
Note that generally, parenthesis using round or square brackets should be kept to a minimum in fiction, and such incidental information is set off with other punctuation, most often commas, if minor, or dashes — for significant chunks of text that include a verb or a comma.
The em-dash (or spaced en-dash) can also be used to mark a sharp external interruption (usually followed by end quote or resuming after a start quote), and should not be confused with the three dot elision mark that indicates where a speaker has trailed off without finishing a thought or giving their conclusions...
Sometimes this is because it is understood, and potentially even completed, by the hearer. Sometimes it may be because they've changed their mind, realized what they were going to say is wrong. Sometimes it is simply an indication that they need time to think... and they will then resume the thought. Like period, the dots should adjoin the word when they mark that part of the word is missing. Like period, it can also adjoin at the end of an utterance (when no actual spoken/written words are omitted). But in contrast with period and apostrophe for just part of a word missing (elided), it is spaced when whole words are missing (elided) — this mostly happens in formal quotation when extraneous details of parenthetical comments are omitted (and square brackets with replacement/summary text can be used instead if needed in order to make a proper sentence).
And it is the three dot pause marker that should be used, not the single point, when you draw out a sentence: never... use... fullstops... The individual words are not fully complete utterances, which is what period indicates. Period! Also, in prose, you don't capitalize after anything but sentence final punctuation (.?!) as (apart from proper nouns/names/titles and 'I') you only capitalize when you are starting a new sentence.
Unfortunately a lot of modern reviewers and publishers, and grammar checkers and teachers, don't take the time to understand the correct usage and logic of such grammar and punctuation, and even promote incorrect usage, and provide incorrect corrections. They also seem to be immune to the prosodic effect and literary intent of careful choice of punctuation. Like Wilde and Kipling, I will sometime agonize for a morning about the precise punctuation and then change my mind again in the afternoon (the earliest attested such anecdote, published in 1884, is attributed to Oscar Wilde: he took out a comma and put it back in again).
Then there are the exclamation mark nazis who would eliminate them all with extreme prejudice. If it is an exclamation, it requires an exclamation mark — and this includes anything ordered or commanded, or delivered with sharp intonation.
I've also had reviews rejected, apparently, for using -ize rather than -ise verbs (fixing these got it approved). This -ize is a productive ending pronounced with the 'z' sound: that is it is used for modern and newly coined words, while ancient -ice and -ise words have roots in other languages (note that in older words the distinction is reflected with -ice/-ise (e.g. device/devise, advice/advise) and some word families have lost either the orthographic distinction or the voicing that normally characterizes the verb (practice/practise, use/use, abuse/abuse, etc., extent/extend, intent/intend), although in speech the missing voicing may still leave marking with extra length for the verb and falling intonation (tone) for the noun. An adjective or noun that doesn't already have its own underlying or cognate verb for how to make/achieve it, can form a verb by adding -ize. When you use something to do something (means rather than result), the noun can be used directly as a verb (he headed the ball to the forward who shouldered past the goalie before kneeing it into the goal).
Coined or coerced words, and new usages in general, are not errors, and are indeed absolutely essential in Science and Science Fiction, to label new inventions, even if Word and Amazon etc. flag them.
Traditionally Published vs Self-Published
So what about the stars?
Of course, the obvious question is exactly what the stars mean in a review or rating. I think of them like grade points, which in turn have a connection to statistical concepts like standard deviation.
Standardized scores (like IQ) are actually scaled to have a nominal mean or median (100) and standard deviation (15). 67-68% of people are then expected to lie within 1 standard deviations (30) of the mean and 95-96% within 2 standard deviations, on the assumption that the distribution of scores (IQ) is relatively normal (bell-shaped).
Marks out of 100 are sometimes standardized with a mean or median of 65 and a standard deviation of 10, so the 55-65-75 Pass/Credit or C/B range corresponds to one standard deviation either way, and grading to the curve tries to achieve ~65% of students in this range (2/3), while 45-55 (Conceded Pass or D) and 75-85 (Distinction or A) can each expect about ~15% (1/7), with 2-4% getting High Distinctions or A+ (1 in 25 to 50), and hopefully <5% of those who complete everything miss the mark and actually fail. Of course, this doesn't account for those that don't even make it to or through the exam. In practice, the pass mark is normally 50, pushing a few more (~8%) from D to C (if normally distributed), and supplementary assessment may also push a student from D to C.
So what about the stars?
***** A or DN+ — maybe 15-20% — if you give five stars out freely, it ceases to mean much!
**** B or CR — maybe 20-25% — this is where most good but not exceptional books will lie
*** C or P — maybe 35-40% — this is fine, passable, but not distinguishable from the crowd
** D or CP — maybe 20-25% — this is poor, but I'll concede it has some redeeming features
* E or F — maybe ~5% — books should have been through enough wringers to avoid this!!!
This effectively converts the percentage marks associated with the grade points to the five star scale, by dividing by 20 and rounding up.
Of course there is a tendency to be lazy regarding the *** books. If there is nothing particularly good or particularly bad about a book, I am disinclined to actually write a detailed review— so may not bother, or may just give it a rating and/or a quick comment. These days I try to give it a rating and may actually look at other reviews and upvote those that I agree with. Reading should be fun, but reviewing is hard work, most especially for those books that lie in the middle with good and bad points.
If a book is getting consistent * and/or ** reviews, then that suggests revision and supplementary assessment is needed! If the reading was hard work, or I did not finish, then (unless there was a genre issue) * or ** is what it will get. If it simply wasn't my kind of book, I might skip the review. But if I can, I will try to express why I found it hard to finish, and identify what kinds of readers I would expect to like and not like the book — and may even rate it ***.
**** is reserved for books I actually liked (in some degree — remember the rounding), while ***** is for those relatively few books that I felt were exceptional, the kind of books I'd like to write — though I'd likely still think there were some things I'd have done differently: it doesn't have to be perfect to get five stars (again, remember the rounding).
Confusingly some rating systems are out of different numbers of stars, e.g. four. In this case I'd divide the percentage by 25 and round up. Actually, I tend to mark out of 10 (for complex work 20) with half marks, and that then multiplies by 10 (resp. 5) to get a percentage, so rating out of 5 in whole numbers doesn't come easy although I am used to it now — and there is still that last minute rounding decision. I tend to rate first, write the review while it percolates, and then adjust at the end if needed (maybe up, if I found positive points to write about).
But I'm think in terms of fractions or decimals or "would be five stars except", and currently I am rating/ranking things for the Aurealis Awards out of 5 (as requested) but with one decimal place — so effectively with 50 shades of grey.
References and Notes
Some people may quibble about the word 'grammo' for 'grammatical error', or more precisely 'confused word error', like the wrong version of their/there/they're.
Google's Ngram Viewer shows 'grammo' has been around since the mid 1800s, but its usage decreased throughout the 20th century, but since 1996 on the increase again.
Google Ngrams are a very useful tool for understanding the chronological (diachronic) usage of words and phrases (Ngrams).
Here is a paper about fixing the 'grammo' in 1997 (it is about a system that corrected such confused words with learned statistics/AI of a kind now ubiquitous in word processors and grammar checkers: the demo system took the form of a macro package for Word):
David M W Powers (1997), Learning and Application of Differential Grammars, CoNLL97, ACL.
• https://aclanthology.org/W97-1011.pdf
Google Scholar is another tool that's worth knowing about to see the impact of scholarly work and track down all an author's publications when writing literature reviews (for non-fiction books and theses) — but that is a whole separate topic (and yes, as both author and reviewer I regard a lit. review and bilbiography/references as essential in non-fiction books: to be taken seriously non-fiction authors need to demonstrate they are familiar with the existing sources of information out there, and ideally critique them; and I've even included footnotes and scientific/technical references/evidence in my fiction, and fictional fables and citations in my non-fiction).
And here is the full quotation and citation for Kipling's serving men:
I keep six honest serving-men
(They taught me all I knew);
Their names are What and Where and When
And How and Why and Who.
I send them over land and sea,
I send them east and west;
But after they have worked for me,
I give them all a rest.
I let them rest from nine till five,
For I am busy then,
As well as breakfast, lunch, and tea,
For they are hungry men.
But different folk have different views;
I know a person small —
She keeps ten million serving-men,
Who get no rest at all!
She sends’em abroad on her own affairs,
From the second she opens her eyes —
One million Hows, two million Wheres,
And seven million Whys!
in The Elephant’s Child (1900)
a Just-so Story
