Let's start a discussion for the weekend : What is a scratch?

Discussion in 'US Coins Forum' started by Insider, Nov 13, 2021.

  1. Insider

    Insider Talent on loan from...

    IMO, a definition should be clear, concise, and extremely focused/applied to the one particular thing being defined. Researching the Internet for a new column raised a question. How should a numismatist define a scratch? What is a scratch?

    One professional organization that I should expect to know the answer and thus make a respectable response defined a scratch this way:

    "A scratch is a type of mechanical damage that has moved or removed metal from the surface of a coin."


    IMHO, this is a very poor definition in part.

    1. What's wrong with it?
    2. Will you post a better definition I can use?

    Thanks!
     
    Mr.Q, imrich, AdamL and 1 other person like this.
  2. Avatar

    Guest User Guest



    to hide this ad.
  3. ToughCOINS

    ToughCOINS Dealer Member Moderator

    The cited definition leaves open the possibility that strike-thrus could be characterized as scratches. It also does a poor job of excluding marks left by tooling a surface or chasing.

    A scratch is a mechanically caused linear (straight and / or curved) depression in the coin's surface, which leaves displaced metal either side of it, above the original surface.
     
    Last edited: Nov 13, 2021
  4. dwhiz

    dwhiz Collector Supporter

    A scratch is what I did to a 1892 Barber 50c TDR:banghead:
    By being careless when removing from a 2 x 2 about 20 years ago.
    Thankfully I don't have a photo, so it never happened?
     
  5. Publius2

    Publius2 Well-Known Member

    This is good as far as it goes but I think it leaves out one or maybe two critical dimensions and those are depth and width. Hairlines meet the above definition but I don't think anyone in the numismatic community considers hairlines to be "scratches". So, somewhere the issue of depth and maybe width needs to be in play here. After all, the difference between a hairline and a scratch is only a matter of degree.

    Off hand, I can't think of a geometric definition of depth and width that, along with the above definition, would narrow it adequately without introducing more confusion. This may be a bit like one Supreme Court justice's definition of obscenity/pornography: "I can't say what it is but I know it when I see it."
     
    Tamaracian and ddddd like this.
  6. alurid

    alurid Well-Known Member

    The removal of metal should be considered a gouge. Linear or circular.
     
  7. expat

    expat Remember you are unique, just like everyone else Supporter

    Quite a difficult subject to describe in a fashion that would be acceptable to the majority.For me hairlines hardly impact the coins surface, scratches are fine like the width of a human hair of any length, getting progressively thicker until I deem them to be a gouge with excess displaced metal at the edges.
    I will be very interested in this discussion as it hopefully progresses. Great subject for a thread @Insider
     
    AdamL and ddddd like this.
  8. ToughCOINS

    ToughCOINS Dealer Member Moderator

    I’d considered mentioning length and width, but the issue with that becomes one of scale. What may be innocuous on a Morgan Dollar could be a major problem on a trime.

    Even trying to define scratches using percentages of coin dimensions may not be appropriate when considering the differential hardness between nickel and gold coinage, for another example.

    Complicated it is, and while I like simplicity inasmuch as it can be achieved, complicated it should stay, I think.
     
    Insider, Tamaracian and tommyc03 like this.
  9. Dynoking

    Dynoking Well-Known Member

    Does a hairline actually displace metal or simply cause contrast?
     
  10. ToughCOINS

    ToughCOINS Dealer Member Moderator

    A hairline technically is a scratch but, by virtue of being shallower and / or narrower, and therefore being less disturbing to the natural appearance of the subject coin, is called by a different name.

    Some augment scratch with hairline, as in "hairline scratch(es)", while others simply refer to them as "hairline(s)".
     
    Tamaracian and Insider like this.
  11. Collecting Nut

    Collecting Nut Borderline Hoarder

    A scratch is what I do when I get poison ivy.
     
    John Skelton likes this.
  12. Casman

    Casman Well-Known Member

    A scratch. I know it when I see it.
     
  13. ksparrow

    ksparrow Coin Hoarder Supporter

    I like ToughCoins definition, but would add "well defined" and "visible to the naked eye".
     
  14. Pickin and Grinin

    Pickin and Grinin Well-Known Member

    Noticeable damage to a coins surface, running in a continuous line,
    Causing unstruck planchet to be exposed.
     
  15. Insider

    Insider Talent on loan from...

    ToughCOINS, posted: "The cited definition leaves open the possibility that strike-thrus could be characterized as scratches. It also does a poor job of excluding marks left by tooling a surface or chasing."

    Yes, "A scratch is a type of mechanical damage that has moved or removed metal from the surface of a coin."

    :rolleyes: This
    covers almost any PMD! So not specific. A very poor :bucktooth: attempt at an important definition.

    Now, let's see some better ones. Think about this: The length of a scratch is not important at all in a good definition. :D
     
    AdamL likes this.
  16. 1stSgt22

    1stSgt22 I'm just me!

    A scratch is damage that could have been done intentionally, accidentally or is the result of stupidity. It happens when a hard object contacts the surface of the object being scratched, for a random distance, with enough force to cause an incused mark!
     
  17. Publius2

    Publius2 Well-Known Member

    Here's a common definition of the verb:

    "score or mark the surface of (something) with a sharp or pointed object."

    Of course, all the terms in this definition are vague. What is sharp? It's all relational. A cat's claw is sharp but not with respect to a bee's stinger. What makes something pointed? All it has to be is somewhat conical in shape with the angle of the tip tending more towards 5 degrees than 80 degrees.

    But we, in a very general sense, agree that a scratch must be considerably longer than it is wide. Call it the L/W ratio. It also should be relatively shallow. Call this the length/depth (L/D) ratio. If not shallow, then we use the term "gouge" to differentiate from "scratch", although the term gouge includes indentations that don't always follow the "longer than wider" dictum.

    So, a scratch should have a large L/W ratio and a large L/D ratio. Would it be a scratch if the L/W ratio were 2? i.e. the length were only two times the width? Most of us would agree that would not constitute a scratch.

    How about L/W=20. Let's put some actual numbers to that: If the width of a mark on a coin were 1/64" (0.0156") then a L/W of 20 would give a scratch length of 0.3125" (5/16"). That's not what most of us would think of as much of a scratch although it would certainly be very visible and maybe many of us might think of it rather as a gouge.

    An L/W of 100 with a width of 1/128" (0.0078") yields a length of 0.7813". Now this is recognizably a scratch over which few would disagree.

    So, a scratch is geometrically characterized by its length divided by width ratio as something in the neighborhood of 50 to 60. The actual number is highly debatable as is this entire exercise, but let's not let that stop us.

    Now to depth. I think most people would agree that the nature of a scratch is to not be deep, whatever that means since the terms are relational. For example, the Grand Canyon is merely a scratch on the earth's surface in that it is long when compared to its width and it is shallow when compared to its length. So what is a proper L/D ratio? Let's take our previous example with a L/W of 100 with a width of 0.0078" and a corresponding length of 0.7813". For this scratch to be a scratch, it needs to be shallow in relation to its length so the L/D ratio needs to be large. How large? If the depth were equal to the width in this example, 0.0078", then the ratio L/W would be 100. It's almost looking like the geometric definition of a scratch is a mark that meets L/D and L/W ratios somewhere in the neighborhood of 50 to 100.

    Let's take a slightly different tack based upon what's been developed so far. How wide does a mark have to be before the unaided 20/20 human eye can even detect it? Well, a quick search on the net tells me that number is about 10 microns or 0.00039". So, if a "scratch" narrower than 0.0004" is not detectable to the unaided human eye, then is it even a scratch? I say no so let's proceed from there. W=0.0004". For a L/W of 50, the scratch length would be 0.02", not much of a scratch at all. For L/W of 100, the length would be 0.04", certainly detectable but we might not call it a scratch but some other kind of a mark. These kinds of extremely narrow scratches are probably more in the category of what numismatists call "hairlines" which are a special category of scratches and not what I think Insider is getting at.

    We could also take a look at the length of the mark in relation to the diameter of the coin. Again, a dimensionless ratio might work in this regard.

    We could work this further looking at the W/D ratio to see if that made any sense but you should have a sense of what I'm getting at here - is there a dimensionless ratio or ratios that more or less captures what we all generally agree constitutes a scratch?

    Of course, we can add non-numerical descriptors to further tighten the definition such as "displaced metal to one or more sides of the scratch and/or at the beginning and end of the scratch."

    And finally, congratulations to anyone who's had the stamina to wade through all this late-night rambling.
     
    Tamaracian and Insider like this.
  18. Jack D. Young

    Jack D. Young Well-Known Member

    And of course there is the die cutters "scratch" which leaves an easy attribution point in several early American coppers:D...
     
    Insider likes this.
  19. bradgator2

    bradgator2 Well-Known Member

    I was told there would be no math
     
    Randy Abercrombie and Insider like this.
  20. Jaelus

    Jaelus The Hungarian Antiquarian Supporter

    We all seem to agree that a hairline is a type of scratch, so then hairlines should be included in the definition. It is a completely separate point to define how the severity of a scratch (including hairlines) affects or does not affect the grade when evaluating a coin.

    For example, certainly there are some lower grade coins that would grade with a minor scratch, depending on the severity, location, originality, etc. Also hairlines aren't evaluated the same way depending on the coin's finish. You don't adjust the definition of scratch to exclude these types of occurrences, that is a separate point.

    In other words a scratch doesn't magically transform into something else for those instances where it doesn't render a coin details. It's just being handled via the grading process. Hairlines are scratches. Period. Don't overthink it.
     
    Last edited: Nov 14, 2021
    Tamaracian, Insider and ToughCOINS like this.
  21. ToughCOINS

    ToughCOINS Dealer Member Moderator

    @Jaelus , I second that . . . Identifying severity comes after identifying the disturbance as a scratch, not in order to do so.

    To take things further, and in line with @Publius2 commentary, we do need a means of differentiating ordinary contact marks from scratches.

    I think that the L/W ratio he mentioned (aspect ratio, in my vernacular), while I discourage people from carrying rulers or dial calipers around with them at coin shows, is a simple means of separating contact marks from scratches. This is bound to be a highly subjective measurement, based on personal preference, on coin, on location, etc., and I don't think there should be an industry standard, as such.
     
    Last edited: Nov 14, 2021
    John Skelton, Insider and Jaelus like this.
Draft saved Draft deleted

Share This Page